\n

Orientation for new MSE in Financial Mathematics s tudents will be held August 12 – August 25.

\nA full orientation schedule for 2015 will be avai lable shortly before the start of orientation.

\nDTSTART;VALUE=DATE:20150812 DTEND;VALUE=DATE:20150813 SEQUENCE:0 SUMMARY:Financial Mathematics Orientation Begins URL:https://engineering.jhu.edu/ams/events/financial-mathematics-orientatio n-begins-2/ X-TAGS;LANGUAGE=en-US:Financial Mathematics END:VEVENT BEGIN:VEVENT UID:ai1ec-5853@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION: DTSTART;VALUE=DATE:20150827 DTEND;VALUE=DATE:20150828 SEQUENCE:0 SUMMARY:Fall Classes Begin URL:https://engineering.jhu.edu/ams/events/fall-classes-begin/ X-TAGS;LANGUAGE=en-US:current students END:VEVENT BEGIN:VEVENT UID:ai1ec-5687@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES;LANGUAGE=en-US:Seminar\,Student Seminar CONTACT: DESCRIPTION:

Given a multi-dimensional data set\, principal component analysis (PCA) is commonly applied to project the data into some low-dimensional s ubspace before subsequent inference. Now suppose two correlated data sets are given\, and they are separately projected by PCA\; we do a Procrustes analysis to compare the separately projected data sets\, and relate the as ymptotic Procrustes fitting error to the Hausdorff distance between the tw o PCA subspaces.

\n\n

DTSTART;TZID=America/New_York:20150324T150000 DTEND;TZID=America/New_York:20150324T160000 LOCATION:Whitehead 304 SEQUENCE:0 SUMMARY:Student Seminar: Cencheng Shen URL:https://engineering.jhu.edu/ams/events/student-seminar-cencheng-shen-2/ END:VEVENT BEGIN:VEVENT UID:ai1ec-5671@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES;LANGUAGE=en-US:Seminar CONTACT: DESCRIPTION:\n

We will present a very ge neral framework for unconstrained stochastic optimization which is based o n standard trust region framework using random models. In particular this framework retains the desirable features such step acceptance criterion\, trust region adjustment and ability to utilize of second order models. We make assumptions on the stochasticity that are different from the typical assumptions of stochastic and simulation-based optimization. In particula r we assume that our models and function values satisfy some good quality conditions with some probability fixed\, but can be arbitrarily bad otherw ise. We will analyze the convergence of this general framework and discuss the requirement on the models and function values. We will will contrast our results with existing results from stochastic approximation literature .

\nWe will then present computational results for several classes o f noisy functions\, including cases when noise is not i.i.d. and dominates the function values\, when it occurs. We will show that our simple framew ork performs very well in that setting\, while standard stochastic methods fail.

\nDTSTART;TZID=America/New_York:20150326T133000 DTEND;TZID=America/New_York:20150326T143000 LOCATION:Hodson 110 SEQUENCE:0 SUMMARY:Seminar: Katya Scheinberg (Lehigh University) URL:https://engineering.jhu.edu/ams/events/seminar-katya-scheinberg-lehigh- university/ END:VEVENT BEGIN:VEVENT UID:ai1ec-5607@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES;LANGUAGE=en-US:Seminar\,Student Seminar CONTACT: DESCRIPTION:

There exists a significant body of work on determining the acquisition number of various graphs when the vertices of those graphs are each initially assigned a unit weight. We stu dy the size of residual set of the path\, star\, complete\, complete bipar tite\, cycle\, and wheel graphs for variations on this initial weighting s cheme\, with the majority of our work focusing on the acquisition number o f randomly weighted graphs. In particular\, we bound the expected acquisit ion number of the $n$-path when n “units” of integral weight\, or chips\, are randomly distributed across its vertices between $0.26n$ and $0.43n$. We then use Azuma’s Lemma to prove that this expected value is tightly con centrated. Additionally\, we offer a non-optimal acquisition protocol algo rithm for the randomly weighted path and compute the expected size of the resultant residual set.

DTSTART;TZID=America/New_York:20150331T150000 DTEND;TZID=America/New_York:20150331T160000 LOCATION:Whitehead 304 SEQUENCE:0 SUMMARY:Student Seminar: Yiguang Zhang URL:https://engineering.jhu.edu/ams/events/student-seminar-yiguang-zhang/ X-TAGS;LANGUAGE=en-US:Yiguang Zhang END:VEVENT BEGIN:VEVENT UID:ai1ec-5672@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES;LANGUAGE=en-US:Seminar CONTACT: DESCRIPTION:\nNetwork int erdiction refers to a game in which an interdictor with limited budget see ks to disrupt the network operation of an interdictee. Interdiction models date back to the early days of Operations Research\, and have found appli cations in drug enforcement optimization\, nuclear smuggling\, and electri cal grid analysis. It is traditional in the literature to use linear progr amming duality for the interdictee’s problem to reformulate these two-leve l problems as bilinear programs.

\nIn this talk\, we study convex re laxations of such bilinear programs. In particular\, we obtain\, in the sp ace of their defining variables\, a linear description of the convex hull of graphs of bilinear functions over the Cartesian product of a general po lytope and a simplex. This result is general and can be applied to a large variety of bilinear programs. For the special case of network interdictio n\, it yields improved linearization constraints that are cognizant of pat hs and cycles of the network. This linearization provides a convex hull de scription of a suitable problem relaxation and we show computationally tha t it leads to significant gap reductions over the traditional linearizatio n of McCormick. We conclude the talk by highlighting applications and exte nsions of the result to complementarity- and cardinality-constrained probl ems.

\nThis talk is based on joint work with Danial Davarnia (UF) an d Mohit Tawarmalani (Purdue).

\n\n

DTSTART;TZID=America/New_York:20150402T133000 DTEND;TZID=America/New_York:20150402T143000 LOCATION:Hodson 110 SEQUENCE:0 SUMMARY:Seminar: Jean-Philippe Richard (University of Florida) URL:https://engineering.jhu.edu/ams/events/seminar-jean-philippe-richard-un iversity-of-florida/ END:VEVENT BEGIN:VEVENT UID:ai1ec-5688@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES;LANGUAGE=en-US:Seminar\,Student Seminar CONTACT: DESCRIPTION:

In thi s talk we consider the problem of determining the presence and location of a static object within an area of interest (AOI) by combining information from multiple sensors. It is important to mention that the terms “object” and “AOI” will be used in a very general sense. It is assumed that there are two classes of sensors: a large sensor capable of searching the entir e AOI and a set of small sensors which (collectively) search only a subset of the AOI. In order to combine information we propose a system identific ation framework based on maximum-likelihood (ML) estimation. This requires collecting several measurements (samples) from each sensor. The ML approa ch allow us to borrow existing convergence and asymptotic normality result s from the literature. While the ideas are somewhat general\, I will give a short illustration using a simple example.

\n\n

In this talk we consider the propagation of values in recursive binary sequences. Such sequences have been heavily studied in th e context of feedback shift registers. Here we prove some results on conve rgence of sequences in terms of greatest common divisors of elements in un derlying delay sets.

DTSTART;TZID=America/New_York:20150407T160000 DTEND;TZID=America/New_York:20150407T170000 LOCATION:Whitehead 304 SEQUENCE:0 SUMMARY:Student Seminar: Karla Hernandez & Heather Gaddy Patsolic URL:https://engineering.jhu.edu/ams/events/student-seminar-karla-hernandez- 3/ END:VEVENT BEGIN:VEVENT UID:ai1ec-5644@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES;LANGUAGE=en-US:Seminar CONTACT: DESCRIPTION:Thi s talk will focus\, in the first place\, on the general setting and use of shape spaces in problems related to computational anatomy. We will show h ow the introduction of large deformation groups equipped with their Rieman nian metrics coupled with tools from geometric measure theory allows to gi ve a ell-posed mathematical formulation of atlas estimation problems on po pulation of curves and surfaces. The second part of the talk will present an extension of this approach for geometric-functional objects.

\n< /p> DTSTART;TZID=America/New_York:20150409T133000 DTEND;TZID=America/New_York:20150409T143000 LOCATION:Hodson 110 SEQUENCE:0 SUMMARY:Seminar: Nicolas Charon (Johns Hopkins University) URL:https://engineering.jhu.edu/ams/events/seminar-nicolas-charon-johns-hop kins-university/ END:VEVENT BEGIN:VEVENT UID:ai1ec-5709@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES;LANGUAGE=en-US:Seminar\,Student Seminar CONTACT: DESCRIPTION:

We s tudy the symplectic reduction of the phase space describing $k$ particles in $\\mathbb{R}^n$ with total angular momentum zero. This corresponds to the singular symplectic quotient associated to the action of the orthogona l group at the zero value of the moment map. We give a description of the ideal of relations of the ring of regular functions of the symplectic quot ient.

\n\n

DTSTART;TZID=America/New_York:20150414T160000 DTEND;TZID=America/New_York:20150414T170000 LOCATION:Whitehead 304 SEQUENCE:0 SUMMARY:Student Seminar: Joshua Cape URL:https://engineering.jhu.edu/ams/events/student-seminar-joshua-cape/ END:VEVENT BEGIN:VEVENT UID:ai1ec-5653@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES;LANGUAGE=en-US:Seminar CONTACT: DESCRIPTION:

We present advances in nonlinear programming th at enable the solution of large-scale problems arising in the control and dispatch of infrastructure systems such as electricity\, natural gas\, and water networks. Our advances involve new strategies to deal with negative curvature and rank deficiencies in a matrix-free setting and the developm ent of scalable numerical linear algebra strategies capable of exploiting embedded structures.

\nDTSTART;TZID=America/New_York:20150416T133000 DTEND;TZID=America/New_York:20150416T143000 LOCATION:Hodson 110 SEQUENCE:0 SUMMARY:Seminar: Victor Zavala (Argonne National Laboratory) URL:https://engineering.jhu.edu/ams/events/seminar-victor-zavala-argonne-na tional-laboratory/ END:VEVENT BEGIN:VEVENT UID:ai1ec-5696@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES;LANGUAGE=en-US:Alumni Event\,Award Ceremony CONTACT:ams_dept@jhu.edu DESCRIPTION:

Reception for AMS Department Alumni\, Faculty and Students\n

Please join us for appetizers and help us congratulate this year’s award winners. We will be announcing this year’s winners of the Joel Dean Excellence in Teaching award\, the Naddor Prize\, the AMS Achievement\, an d the Mathematical Modeling Award.

DTSTART;TZID=America/New_York:20150417T160000 DTEND;TZID=America/New_York:20150417T180000 LOCATION:Whitehead 304 SEQUENCE:0 SUMMARY:Alumni Reception & Award Ceremony URL:https://engineering.jhu.edu/ams/events/alumni-reception-award-ceremony/ END:VEVENT BEGIN:VEVENT UID:ai1ec-5689@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES;LANGUAGE=en-US:Seminar\,Student Seminar CONTACT: DESCRIPTION:\nA canonical problem i n graph mining is the detection of dense communities. This problem is exac erbated for a graph with a large order and size — the number of vertices a nd edges — as many community detection algorithms scale poorly. In this wo rk we propose a novel framework for detecting active communities that cons ist of the most active vertices in massive graphs. The framework is applic able to graphs having billions of vertices and hundreds of billions of edg es. Our framework utilizes a parallelizable trimming algorithm based on a locality statistic to filter out inactive vertices\, and then clusters the remaining active vertices via spectral decomposition on their similarity matrix. We demonstrate the validity of our method with synthetic Stochasti c Block Model graphs\, using Adjusted Rand Index as the performance metric . We further demonstrate its practicality and efficiency on a real-world H yperlink Web graph consisting of over 3.5 billion vertices and 128 billion edges.

\nDTSTART;TZID=America/New_York:20150421T150000 DTEND;TZID=America/New_York:20150421T160000 LOCATION:Whitehead 304 SEQUENCE:0 SUMMARY:Student Seminar: Heng Wang URL:https://engineering.jhu.edu/ams/events/student-seminar-heng-wang-2/ X-TAGS;LANGUAGE=en-US:Heng Wang END:VEVENT BEGIN:VEVENT UID:ai1ec-5706@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES;LANGUAGE=en-US:Seminar CONTACT: DESCRIPTION:

We give a brief overview of the history of the Monte Carlo method for the nu merical solution of partial differential equations (PDEs) focusing on the Feynman-Kac formula for the probabilistic representation of the solution of the PDEs. We then take the example of solving the linearized Poisson-Boltzmann equation to compare and contrast standard deterministic numerical approac hes with the Monte Carlo method. Monte Carlo methods have always been popu lar due to the ease of finding computational work that can be done in para llel. We look at how to extract parallelism from Monte Carlo methods\, and some newer ideas based on Monte Carlo domain decomposition that extract e ven more parallelism. In light of this\, we look at the implications of us ing Monte Carlo to on high-performance architectures and algorithmic resil ience.

\n**VIEW SLIDES FROM THIS PRESENTATION**

\n

DTSTART;TZID=America/New_York:20150423T133000 DTEND;TZID=America/New_York:20150423T143000 LOCATION:Hodson 110 SEQUENCE:0 SUMMARY:Seminar: Michael Mascagni (Florida State University) URL:https://engineering.jhu.edu/ams/events/seminar-michael-mascagni-florida -state-university/ END:VEVENT BEGIN:VEVENT UID:ai1ec-5674@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES;LANGUAGE=en-US:Outside Seminar\,Seminar CONTACT: DESCRIPTION:

**Abstract:**

\nWe give algorithms for
regression for a wide class of M-Estimator loss functions. These generali
ze l_p-regression to fitness measures used in practice such as the Huber m
easure\, which enjoys the robustness properties of l_1 as well as the smoo
thness properties of l_2. For such estimators we give the first input spar
sity time algorithms. Our techniques are based on the sketch and solve par
adigm. The same sketch works for any M-Estimator\, so the loss function ca
n be chosen after compressing the data.

\nJoint work with Ken Clarkso
n.

\n** Bio:**

\nDavid Woodruff received his Ph.D.
from MIT in 2007 and has been a research scientist at IBM Almaden since th
en. His research interests are in big data\, including communication compl
exity\, compressed sensing\, data streams\, machine learning\, and numeric
al linear algebra. He is the author of the book “Sketching as a Tool for N
umerical Linear Algebra”. He received best paper awards in STOC and PODS\,
and the Presburger award.

Host: Vova Braverman

DTSTART;TZID=America/New_York:20150428T133000 DTEND;TZID=America/New_York:20150428T143000 LOCATION:Malone 228 SEQUENCE:0 SUMMARY:Joint Seminar with CS: David Woodruff (IBM) URL:https://engineering.jhu.edu/ams/events/joint-seminar-with-cs-david-wood ruff/ END:VEVENT BEGIN:VEVENT UID:ai1ec-5714@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES;LANGUAGE=en-US:Dept Event CONTACT: DESCRIPTION:As a chance to relax before finals/eat some good food/talk t o Allen\, we would like to invite everyone to this semester’s AMS Picnic!< /p>\n

The picnic will be on Sunday May 3 from 12 p.m. – 3 p.m in front o f Whitehead. Food will be provided\, but feel free to bring your favorite snack or drink to share with the department (the dedicated student can sho w that the amount of food brought to the picnic is inversely proportional to the amount of graduate student hunger).

DTSTART;TZID=America/New_York:20150503T120000 DTEND;TZID=America/New_York:20150503T150000 LOCATION:Grassy Lawn between Whitehead and Brody Commons SEQUENCE:0 SUMMARY:Spring AMS Department Picnic URL:https://engineering.jhu.edu/ams/events/spring-ams-department-picnic/ X-TAGS;LANGUAGE=en-US:grads\,masters\,undergrads END:VEVENT BEGIN:VEVENT UID:ai1ec-5713@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES;LANGUAGE=en-US:HUSAM CONTACT:HUSAM.JHU@gmail.com\; https://www.facebook.com/events/1047801891916 611/ DESCRIPTION:John Urschel is a football player for the Baltimore Ravens\;
this past year he played in 11 games. He is also an accomplished

\nm
athematician\, the first-author of a paper in the Journal of Computational
Mathematics\, in which he developed fast numerical methods of computing t
he eigenvector associated with the second smallest eigenvalue of a graph L
aplacian. How does his professional football experience relate to the esot
eric world of cutting-edge mathematical research?

Join HUSAM in we lcoming John to share his fascinating intersection of the gridiron and num erical analysis at the highest levels.

DTSTART;TZID=America/New_York:20150504T180000 DTEND;TZID=America/New_York:20150504T190000 LOCATION:Hodson 110 SEQUENCE:0 SUMMARY:HUSAM: John Urschel (Baltimore Ravens/Mathematician) URL:https://engineering.jhu.edu/ams/events/husam-john-urschel-baltimore-rav ensmathematician/ X-TAGS;LANGUAGE=en-US:HUSAM END:VEVENT BEGIN:VEVENT UID:ai1ec-5690@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES;LANGUAGE=en-US:Seminar\,Student Seminar CONTACT: DESCRIPTION:In this talk we introduce the semi-supervised clustering problem. Then\, we explicate the model-based a pproach with some commentary on initialization schemes using other semi-su pervised clustering algorithms (i.e. constrained K-means++). Next\, we sk etch a proof for an improved approximation bound for constrained K-means++ . Finally\, we apply our methods to two applications: vertex nomination a nd worm brain clustering.

\nDTSTART;TZID=America/New_York:20150505T150000 DTEND;TZID=America/New_York:20150505T160000 LOCATION:Whitehead 304 SEQUENCE:0 SUMMARY:Student Seminar: Jordan Yoder URL:https://engineering.jhu.edu/ams/events/student-seminar-jordan-yoder/ END:VEVENT BEGIN:VEVENT UID:ai1ec-5852@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES;LANGUAGE=en-US:PhD CONTACT: DESCRIPTION: DTSTART;TZID=America/New_York:20150819T083000 DTEND;TZID=America/New_York:20150819T170000 LOCATION:Whitehead 304 SEQUENCE:0 SUMMARY:PhD Introductory Exam URL:https://engineering.jhu.edu/ams/events/phd-introductory-exam-2/ X-TAGS;LANGUAGE=en-US:Intro Exam\,Orientation\,PhD END:VEVENT BEGIN:VEVENT UID:ai1ec-5976@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION: DTSTART;TZID=America/New_York:20150826T143000 DTEND;TZID=America/New_York:20150826T163000 SEQUENCE:0 SUMMARY:Teaching Assistant Orientation: Meet in Hodson 316 URL:https://engineering.jhu.edu/ams/events/teaching-assistant-orientation-m eet-in-hodson-316/ END:VEVENT BEGIN:VEVENT UID:ai1ec-5974@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION: DTSTART;TZID=America/New_York:20150827T133000 DTEND;TZID=America/New_York:20150827T143000 SEQUENCE:0 SUMMARY:Seminar: Get to Know You @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-get-to-know-you/ END:VEVENT BEGIN:VEVENT UID:ai1ec-5959@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:

A GENERAL THEORY FOR COMPUTING ATTRACTIVE REPRESENTATIONS ON NONCONVEX OPTIMIZATION PROBLEMS

\nMore than one mathematical repres entation can accurately depict a decision problem. Success in obtaining op timal solutions\, however\, often depends upon the formulation selected. S ince challenging nonconvex optimization problems are typically solved by u sing linear programming relaxations as tools to compute bounds for elimina ting inferior solutions\, “attractive” representations tend to be characte rized by the accuracy of their relaxations. This importance of relaxation strength is well documented within the Operations Research literature\, wh ere numerous authors have suggested methods for acquiring strength. The po sed methods are often problem dependent\, relying on the exploitation of s pecific structures.

\nThis talk presents a general theory for derivi ng representations with tight relaxations. The fundamental idea is to reca st a given problem into higher-dimensional spaces by automatically generat ing auxiliary variables and constraints. Strength is garnered via suitable mathematical identities. The talk begins with an introduction to the impo rtance of relaxation strength\, and then highlights contributions and chal lenges relative to the progressively more general families of mixed-binary \, mixed-discrete\, and general nonconvex programs. Ongoing research is di scussed.

DTSTART;TZID=America/New_York:20150903T133000 DTEND;TZID=America/New_York:20150903T143000 SEQUENCE:0 SUMMARY:Seminar: Warren Adams (Clemson University) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-warren-adams-clemson-uni versity-whitehead-304/ END:VEVENT BEGIN:VEVENT UID:ai1ec-5941@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:SUBGROUP-BASED ADAPTIVE (SUBA) ENRICHMENT DESIGNS FOR MULT-A RM BIOMARKER TRIALS

\nTargeted therapies based on biomarker profilin g are becoming a mainstream direction of cancer research and treatment. De pending on the expression of specific prognostic biomarkers\, targeted the rapies assign different cancer drugs to subgroups of patients even if they are diagnosed with the same type of cancer by traditional means\, such as tumor location. For example\, Herceptin is only indicated for the subgrou p of patients with HER2+ breast cancer\, but not other types of breast can cer. However\, subgroups like HER2+ breast cancer with effective targeted therapies are rare and most cancer drugs are still being applied to large patient populations that include many patients who might not respond or be nefit. Also\, the response to targeted agents in humans is usually unpredi ctable. To address these issues\, we propose SUBA\, subgroup-based adaptiv e designs that simultaneously search for prognostic subgroups and allocate patients adaptively to the best subgroup-specific treatments throughout t he course of the trial. The main features of SUBA include the continuous r eclassification of patient subgroups based on a random partition model and the adaptive allocation of patients to the best treatment arm based on po sterior predictive probabilities. We compare the SUBA design with three al ternative designs including equal randomization\, outcome-adaptive randomi zation and a design based on a probit regression. In simulation studies we find that SUBA compares favorably against the alternatives.

DTSTART;TZID=America/New_York:20150910T133000 DTEND;TZID=America/New_York:20150910T143000 SEQUENCE:0 SUMMARY:Seminar: Yanxun Xu (Johns Hopkins University) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-yanxun-xu-johns-hopkins- university/ END:VEVENT BEGIN:VEVENT UID:ai1ec-5983@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:THE GENERAL SETTING FOR SHAPE DEFORMATION ANALYSIS

\nI will define a unified setting for shape registration and LDDMM methods fo r shape analysis\, using optimal control theory\, and give the Hamiltonian geodesic equations associated to a smooth enough reproducing kernel. I wi ll then give several applications of this framework\, such as fibered shap es (for muscles)\, and the addition of constraints for the simultaneous st udy of multiple interacting shapes.

DTSTART;TZID=America/New_York:20150917T133000 DTEND;TZID=America/New_York:20150917T143000 SEQUENCE:0 SUMMARY:Seminar: Sylvain Arguillere (Johns Hopkins University) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-sylvain-arguillere-johns -hopkins-university-whitehead-304/ END:VEVENT BEGIN:VEVENT UID:ai1ec-5855@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES;LANGUAGE=en-US:WSE CONTACT: DESCRIPTION: DTSTART;TZID=America/New_York:20150918T120000 DTEND;TZID=America/New_York:20150918T140000 SEQUENCE:0 SUMMARY:WSE Annual Fall Picnic URL:https://engineering.jhu.edu/ams/events/wse-annual-fall-picnic-2/ X-TAGS;LANGUAGE=en-US:current students\,faculty\,staff END:VEVENT BEGIN:VEVENT UID:ai1ec-5938@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:CHALLENGES IN GRAPH-BASED MACHINE LEARNING AND ROBUSTIFYING DATA GRAPHS WITH SCALABLE LOCAL SPECTRAL METHODS

\nGraphs are very p opular ways to model data in many data analysis and machine learning appli cations\, but they can be quite challenging to work with\, especially when they are very sparse\, as is typically the case. We will discuss challen ges we have encountered in working with large sparse graphs in machine lea rning and data analysis applications and in particular in the construction of these graphs\, e.g.\, with various sorts of popular nearest neighbor r ules applied to feature vectors. In our experience\, many properties of t he constructed graphs are very sensitive to seemingly-minor and often-igno red aspects of the graph construction process. This should suggest cautio n in using popular algorithmic and statistical tools\, e.g.\, popular nonl inear dimensionality reduction methods\, in trying to extract insight from those constructed graphs. We will also describe recent results on using local spectral methods to robustify this graph construction process. Loca l spectral methods use locally-biased random walks\, they have had several remarkable successes in worst-case algorithm design as well as in analyzi ng the empirical properties of large social and information networks\, and they are an example of a worst-case approximation algorithm that implicit ly but exactly implements a form of statistical regularization. Informall y\, the reason for the successes of these methods in robustifying graph co nstruction is that these local random walks provide a regularized or stabl e version of an eigenvector\, and initial results on using these ideas to robustify the graph construction process are promising.

DTSTART;TZID=America/New_York:20150924T133000 DTEND;TZID=America/New_York:20150924T143000 SEQUENCE:0 SUMMARY:Seminar: Michael Mahoney (University of California\, Berkeley) @ Wh itehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-michael-mahoney-uc-berke ley/ END:VEVENT BEGIN:VEVENT UID:ai1ec-5702@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES;LANGUAGE=en-US:Seminar CONTACT: DESCRIPTION:GAUGE DUALITY AND LOW-RANK SPECTRAL OPTIMIZATION

\n\n

Gauge functions significantly generalize the notion of a norm\, and

\ngauge optimization is the class of problems for finding the eleme nt of

\na convex set that is minimal with respect to a gauge. These< /p>\n

conceptually simple problems appear in a remarkable array of

\napplications. Their structure allows for a special kind of duality

\nframework that can lead to new algorithmic approaches to challenging< /p>\n

problems. Low-rank spectral optimization problems that arise in tw o

\nsignal-recovery application\, phase retrieval and blind deconvol ution\,

\nillustrate the benefits of the approach.

DTSTART;TZID=America/New_York:20151001T133000 DTEND;TZID=America/New_York:20151001T143000 SEQUENCE:0 SUMMARY:Seminar: Michael Friedlander (University of California\, Davis) @ W hitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-michael-friedlander-univ ersity-of-california-davis/ END:VEVENT BEGIN:VEVENT UID:ai1ec-5937@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES;LANGUAGE=en-US:Seminar CONTACT: DESCRIPTION:Change Point Inference for Time-varying Erdos-Renyi Graphs\n

We investigate a model of an Erdos-Renyi graph\, where the edges ca n be in a present/absent state. The states of each edge evolve as a Markov chain independently of the other edges\, and whose parameters exhibit a c hange-point behavior in time. We derive the maximum likelihood estimator f or the change-point and characterize its distribution. Depending on a meas ure of the signal-to-noise ratio present in the data\, different limiting regimes emerge. Nevertheless\, a unifying adaptive scheme can be used in p ractice that covers all cases.We illustrate the model and its flexibility on US Congress voting patterns using roll call data.

DTSTART;TZID=America/New_York:20151008T133000 DTEND;TZID=America/New_York:20151008T143000 SEQUENCE:0 SUMMARY:Seminar: George Michailidis (University of Florida) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-george-michailidis-unive rsity-of-florida/ END:VEVENT BEGIN:VEVENT UID:ai1ec-6073@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES;LANGUAGE=en-US:HUSAM CONTACT: DESCRIPTION:Lies\, Deceit\, and Misrepresentation: The Distortion of Sta tistics in America

\nH.G. Wells once said “Statistical thinking will one day be as necessary for efficient citizenship as the ability to read and write.” The widespread use of statistics plays an influential role in persuading public opinion. As such\, statistical literacy is necessary for members of society to critically evaluate the bombardment of charts\, pol ls\, graphs\, and data that are presented on a daily basis. However\, what often passes for “statistical” calculations and discoveries need to be ta ken with a grain of salt. This talk will examine the applications of stati stics in American media and give examples of where statistics has been gro ssly misused.

\nThe talk will begin at 7pm in Hodson 110\, with refr eshments being served at 6:30. A flyer for the event is attached and a lin k to RSVP on the Facebook page is here: https://www.facebook.com/events/959982947374497/< /a>.

DTSTART;TZID=America/New_York:20151021T183000 LOCATION:Hodson 110 SEQUENCE:0 SUMMARY:HUSAM: Dr. Talitha Williams\, Harvey Mudd College URL:https://engineering.jhu.edu/ams/events/husam-dr-talitha-williams-harvey -mudd-college/ X-TAGS;LANGUAGE=en-US:HUSAM X-INSTANT-EVENT:1 END:VEVENT BEGIN:VEVENT UID:ai1ec-5989@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:**Title**: *Information Theoretic Intuitions
about Some Estimation Problems in Speech Recognition*

**Abstract**:

Automatic speech recognition (ASR) systems com pose probabilistic models of numerous kinds to transcribe a spoken utteran ce into a sequence of written words. The so called acoustic model aims to compute the probability of each sound category label\, such as a phoneme or an allophone\, given a short segment of speech. The parameters of the acoustic model are typically estimated to maximize discrimination between the correct and incorrect sound categories on some labeled “training” samp les—specifically to maximize a mutual information. The inventory of sound categories (allophones) the acoustic model is trained to discriminate is also determined from data. This is typically done using decision trees to recursively divide all acoustic samples of a phoneme\, based on the phone tic context of the sample\, into maximally homogeneous subsets—specificall y\, subsets that minimize a conditional entropy.

\nTwo recent advanc es\, one each in acoustic model estimation and in the creation of phonetic decision trees\, will be described\, beginning with the information theor etic intuitions behind the changes we made to currently used methods. The first replaces the maximization of mutual information with minimization o f a related conditional entropy\, which turns out to be advantageous for s emi-supervised training of acoustic models\, i.e. when some samples have m issing labels. The second investigates an alternative to random forests by developing multiple decisions trees in a deterministic manner\; it maxi mizes diversity by minimizing mutual information between the leaves assign ed by the multiple trees to each training sample.

\nThe presentation will have a tutorial flavor for a non-speech-technology audience\, and wi ll not present any new results in mathematics or statistics\; just some ma thematical intuitions about estimation techniques that appear to improve s peech recognition performance on benchmark data sets.

\n**Bios
ketch**:

Sanjeev Khudanpur (PhD 1997\, Electrical Engineeri ng\, University of Maryland) is an Associate Professor in the Departments of Electrical and Computer Engineering and of Computer Science\, the Actin g Director of the Center for Language and Speech Processing\, and a foundi ng affiliate of the Human Language Technology Center of Excellence\, all i n The Johns Hopkins University. His interests are in the application of s tatistical methods to speech and text processing\, and to other engineerin g problems involving time-series data. His office is in Hackerman Hall\, the Homewood campus building with the surgical robots!

DTSTART;TZID=America/New_York:20151022T133000 DTEND;TZID=America/New_York:20151022T143000 SEQUENCE:0 SUMMARY:Seminar: Sanjeev Khudanpur (Johns Hopkins University) @ Whitehead 3 04 URL:https://engineering.jhu.edu/ams/events/seminar-sanjeev-khudanpu-johns-h opkins-university-whitehead-304/ END:VEVENT BEGIN:VEVENT UID:ai1ec-5960@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Recent Results on Polynomial Optimization Problems

\nP olynomial optimization problems\, as the name suggests\, are optimization problem where the objective function as well as the constraints are descri bed by polynomials. Such problems have acquired increased interest to som e degree because of applications in engineering and science\, where constr aints arise because of physics\, and also because of increased theoretical understanding. In this talk I will focus on two topics where I am workin g\, the CDT (Celis Dennis Tapia) problem\, which concerns the solution of a system of quadratic inequalities over R^n\, and mixed-integer polynomial optimization problems over graphs with structural sparsity\, i.e. low tre ewidth. We will describe our results\, but also we will discuss how these problems relate to classical problems in various branches of mathematics.

DTSTART;TZID=America/New_York:20151029T133000 DTEND;TZID=America/New_York:20151029T143000 SEQUENCE:0 SUMMARY:Goldman Lecture Series: Daniel Bienstock (Columbia University) @ Kr ieger 205 URL:https://engineering.jhu.edu/ams/events/seminar-daniel-bienstock-columbi a-university-whitehead-304/ END:VEVENT BEGIN:VEVENT UID:ai1ec-5975@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Graduate Research Opportunities in AMS

\nThis seminar will familiarize Master’s and PhD students from AMS or other WSE departmen ts with the research performed by the AMS faculty. It will be composed of a research overview presented by Professor Laurent Younes and consisting of snap-shot descriptions of research projects currently underway as well as others that are ripe for students to tackle immediately\, followed by a discussion session between faculty and students.

\n\n

Research Opportunities in Applied Mathematics and Statistic s

DTSTART;TZID=America/New_York:20151105T133000 DTEND;TZID=America/New_York:20151105T143000 SEQUENCE:0 SUMMARY:Seminar: The AMS Faculty URL:https://engineering.jhu.edu/ams/events/seminar-laurent-younes-johns-hop kins-university/ END:VEVENT BEGIN:VEVENT UID:ai1ec-5734@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES;LANGUAGE=en-US:Seminar CONTACT: DESCRIPTION:Scaling and Generalizing Variational Inference

\n\n

Latent variable models have become a key tool for the modern statisti cian\, letting us express complex assumptions about the hidden structures that underlie our data. Latent variable models have been successfully appl ied in numerous fields.

\nThe central computational problem in laten t variable modeling is posterior inference\, the problem of approximating the conditional distribution of the latent variables given the observation s.

\nPosterior inference is central to both exploratory tasks and pr edictive tasks. Approximate posterior inference algorithms have revolutio nized Bayesian statistics\, revealing its potential as a usable and genera l-purpose language for data analysis.

\nBayesian statistics\, howeve r\, has not yet reached this potential.

\nFirst\, statisticians and scientists regularly encounter massive data sets\, but existing approximat e inference algorithms do not scale well.

\nSecond\, most approximat e inference algorithms are not generic\; each must be adapted to the speci fic model at hand.

\nIn this talk I will discuss our recent research on addressing these two limitations. I will describe stochastic variatio nal inference\, an approximate inference algorithm for handling massive da ta sets. I will demonstrate its application to probabilistic topic models of text conditioned on millions of articles. Then I will discuss black bo x variational inference. Black box inference is a generic algorithm for a pproximating the posterior. We can easily apply it to many models with li ttle model-specific derivation and few restrictions on their properties. I will demonstrate its use on a suite of nonconjugate models of longitudin al healthcare data.

\n\n

Biography:

\nDavid Blei is a P rofessor of Statistics and Computer Science at Columbia University\, and a member of the Columbia Data Science Institute. His research is in statis tical machine learning\, involving probabilistic topic models\, Bayesian n onparametric methods\, and approximate posterior inference algorithms for massive data. He works on a variety of applications\, including text\, im ages\, music\, social networks\, user behavior\, and scientific data. Dav id has received several awards for his research\, including a Sloan Fellow ship (2010)\, Office of Naval Research Young Investigator Award (2011)\, P residential Early Career Award for Scientists and Engineers (2011)\, Blava tnik Faculty Award (2013)\, and ACM-Infosys Foundation Award (2013).

\nDTSTART;TZID=America/New_York:20151112T133000 DTEND;TZID=America/New_York:20151112T143000 SEQUENCE:0 SUMMARY:Seminar: David Blei (Columbia University) @ Krieger 205 URL:https://engineering.jhu.edu/ams/events/seminar-david-blei-columbia-univ ersity/ END:VEVENT BEGIN:VEVENT UID:ai1ec-5645@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:

Algebraic and Geometric ideas in the theory of Linear Optimi zation”

\nAbstract: Linear optimization is undeniably a central tool of applied mathematics with applications in a wide

\nrange of topic s\, from statistical regression to image processing. The theory of linear optimization has many

\nbeautiful geometric and algebraic topics an d it is still a source of many fascinating mathematical open problems.

\n\n

In this talk I will present several advances from the past 10 years in the theory of linear optimization.

\nThese results inc lude new results on the complexity of the simplex method\, the structure of central

\npaths of interior point methods\, and about the geometr y of some less well-known iterative techniques.

\nOne interesting fe ature of these new theorems is that they connect this very applied algorit hmic field with

\nseemingly far away “pure” topics like algebraic g eometry\, differential geometry\, and combinatorial topology.

\n\n

This panoramic talk is geared for students and the non-expert facul ty member. I will summarize work by many

\nauthors\, including resul ts that are my own joint work with subsets of the following people A. Basu \, J. Haddock\,

\n- \n
- Junod\, S. Klee\, B. Sturmfels\, and C. Vi nzant. \n

**Comparative Effectiveness Research of Env
ironmental Exposures: Connecting the Dots with Big Data**

\n

Comparative effectiveness research increasingly depends on the an alysis of a rapidly expanding universe of observational data made possible by the growing integration of administrative claims data (e.g. Medicare o r SEER-Medicare claims) with environmental health exposures (e.g. emission s from power plants\, air pollution for monitoring stations)\, with survey and census data (e.g. population demographics).

\nWe are interested in addressing questions that attempt to connect the dots between environm ental exposures and human health\, such as: Can increased noise levels nea r airports cause higher rates of cardiovascular disease or stroke? Do even moderate increases in air pollution from sources such as automobiles and industrial smokestacks have a measurable effect on a community’s death rat e? What are the most likely causes of hospitalizations during heat waves?< /p>\n

Development of statistical methods is needed to be able to handle large\, messy data sets\, integrate them\, and extract meaningful conclusi ons. In this talk we will review some of these tatistical methods aimed at making causal inferences on the effectiveness of environmental interventi ons with such large observational data structures.

\n\n

\n

Francesca Dominici is a Professor in the Department of Biostatistics at the Harvard School of Public Health and the Senior Associate Dean for Research. Dr. Dominici received her Ph.D. i n Statistics from the University of Padua\, Italy in 1997. During her PhD\ , she spent two years as a visiting PhD student at Duke University\, NC\, USA. In 1997 she went to the Bloomberg School of Public Health as a post- doctoral fellow. In 1999 she was appointed Assistant Professor at the Bloo mberg School of Public Health and in 2007 she was promoted to Full Profess or with Tenure. In 2009 she moved to Harvard School of Public Health as a tenured Professor of Biostatistics\, was appointed Associate Dean of Infor mation Technology in 2010\, and Senior Associate Dean for Research in 2013 .

\nDr. Dominici’s research has focused on the development of statis tical methods for the analysis of large observational data with the ultima te goal of addressing important questions in environmental health science\ , health related impacts of climate change\, and comparative effectiveness research. She is an expert in Bayesian methods\, longitudinal data analys is\, confounding adjustment\, causal inference\, and Bayesian hierarchical models. She has extensive experience on the development of statistical m ethods and their applications to environmental epidemiology\, implementati on science and health policy\, outcome research and patient safety\, and c omparative effectiveness research.

\n\n

**Research**

Dr. Dominici has authored more than 120 peer-reviewed publica
tions. She is the PI\, together with Dr. Xihong Lin\, of a NCI P01 project
entitled “**Statistical Informatics for Cancer Research**” (
http://ww
w.hsph.harvard.edu/statinformatics/index.html). She is the PI of a Pr
oject called “**A National Study to Assess Susceptibility\, Vulnerab
ility and Effect Modification of Air Pollution Health Risks**” as p
art of the Harvard EPA Center entitled **“Air Pollution Mixtures: He
alth Effects Across Life Stages**” (PI: Dr. Koutrakis) She is also
the PI of several EPA/NIH/HEI funded projects aimed at developing statisti
cal methods and conducting nation-wide epidemiological studies on the heal
th effects of air pollution. Most recently\, she has become more involved
in comparative effectiveness research collaborating with investigators at
Dana Farber Cancer Institute. With her colleagues she is developing statis
tical methods for causal inference and propensity score matching to compar
e health care delivery systems in end of life cancer\, with a special focu
s on glioblastoma and pancreatic cancer. Dr. Dominici also oversees the ma
nagement and the analysis of several administrative databases\, including
Part A CMS files and SEER-Medicare\, which are linked to air pollution and
weather and socioeconomic data.

\n

**Education and M
entoring**

Dr. Dominici is teaching the course Bio249 entit led “Bayesian Methodology in Biostatistics” at HSPH. Previously she taught Analysis of Longitudinal Data\, and Multilevel Statistical Models while a faculty member at Johns Hopkins University. She has been the primary advi sory of 9 PhD students and 13 post-doctoral fellows. She is a passionate m entor of junior faculty.

\n\n

**Diversity**

Dr. Dominici is committed to diversity. Together with Dr. Linda P. Fr
ied (now Dean of the Mailman School of Public Health at Columbia Universit
y)\, she has co-chaired the University Committee of the Status of Women at
Johns Hopkins University. From this experience she wrote a paper entitled
“So Few Women Leaders” *Academe*\, *July-August 2009*” (http:
//www.aaup.org/article/so-few-women-leaders – .Ubx4SZWQma4). In 2009\,
she was awarded the Diversity Recognition Award by the President of Johns
Hopkins University. Recently\, she has been giving lectures and moderated
panel discussions on work-family balance across Harvard (see http
://news.harvard.edu/gazette/story/2012/11/having-it-all-at-harvard/).
Currently she is the chair (with Dr. Burleigh) of the University Committee
for the Advancement of Women Faculty at HSPH.

\n

**A
dministration**

In her role as Associate Dean of Informatio n Technology\, Dr. Dominici has led new initiatives at HSPH regarding rese arch computing. More specifically\, she led a MOU between our school and research computing (RC) facility at the Faculty of Arts and Science (FASRC ) http://rc.fas.harvard.edu/ enab ling HSPH faculty to access the FAS computing facilities. HSPH faculty are treated equally to FAS faculty in terms of priority of access\, ticket tu rn-around\, and access to shared facilities and shared licenses. See http://rc.fas.harvard.edu/hs ph-overview/ for details.

\n\n

**Service**\n

Dr. Dominici has served on a number of National Academies’ committe es\, including the Committee on Research Direction in Human Biological Eff ects of Low Level Ionizing Radiation\; the Committee on Gulf War and Healt h: Review of the Medical Literature Relative to Gulf War Veterans’ Health\ ; the Committee to Review the Federal Response to the Health Effects Assoc iated with the Gulf of Mexico Oil Spill\; the Committee on Secondhand Smok e Exposure and Acute Coronary Events\; the Committee to Review ATSDR’s Gre at Lakes Report\; the Committee on Making Best Use of the Agent Orange Exp osure Reconstruction Model\; the Committee on Gulf War and Health\; the Co mmittee to Assess Potential Health Effects from Exposures to PAVE PAWS Low -Level Phased-array Radiofrequency Energy\; and the Committee on the Utili ty of Proximity-Based Herbicide Exposure Assessment in Epidemiologic Studi es of Vietnam Veterans.

\nDr. Dominici has received numerous recogni tions\, including the Florence Nightingale David award\, sponsored jointly by the Committee of Presidents of Statistical Societies and Caucus for Wo men in Statistics 2015\; Mathematics for Planet Earth Award Lecture\, host ed by the Statistical and Applied Mathematical Sciences Institute (SAMSI) 2013\; Diversity Recognition Award\, Johns Hopkins University\, 2009\; Myr to Lefkopoulou Distinguished Lectureship Award\, Department of Biostatisti cs\, Harvard School of Public Health\, 2007\; Gertrude Cox Award\, Washing ton DC Chapter of the American Statistical Association and RTI Internation al\, 2007\; Mortimer Spiegelman Award\, Statistics Section of the American

\nPublic Health Association\, 2006\; Dean’s Lecture\, Bloomberg Sch ool of Public Health\, 2007\; and an Invitation to Address the Royal Stati stical Society\, London\, UK\, 2002.

\nShe is a member of numerous p rofessional societies\, including the American Statistical Association\, t he International Biometric Society\, and the International Society for Env ironmental Epidemiology. She is the Senior Editor of Chapman & Hall/CRC Te xts in Statistical Science Series and Associate Editor of the Journal of t he Royal Statistical Society.

\nDTSTART;TZID=America/New_York:20151203T133000 DTEND;TZID=America/New_York:20151203T143000 SEQUENCE:0 SUMMARY:Wierman Lecture Series: Francesca Dominici (Harvard University) @ A rellano Theater URL:https://engineering.jhu.edu/ams/events/seminar-wierman-lecture-series-f rancesca-dominici-harvard-university-whitehead-304/ END:VEVENT BEGIN:VEVENT UID:ai1ec-6099@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:

Title: Feature allocations\, probability functions\, and pai ntboxes

\nAbstract:

\nClustering involves placing entities int o mutually exclusive categories. We wish to relax the requirement of mutua l exclusivity\, allowing objects to belong simultaneously to multiple clas ses\, a formulation that we refer to as “feature allocation.” The first st ep is a theoretical one. In the case of clustering the class of probabilit y distributions over exchangeable partitions of a dataset has been charact erized (via exchangeable partition probability functions and the Kingman p aintbox). These characterizations support an elegant nonparametric Bayesia n framework for clustering in which the number of clusters is not assumed to be known a priori. We establish an analogous characterization for featu re allocation\; we define notions of “exchangeable feature probability fun ctions” and “feature paintboxes” that lead to a Bayesian framework that do es not require the number of features to be fixed a priori. The second ste p is a computational one. Rather than appealing to Markov chain Monte Carl o for Bayesian inference\, we develop a method to transform Bayesian metho ds for feature allocation (and other latent structure problems) into optim ization problems with objective functions analogous to K-means in the clus tering setting. These yield approximations to Bayesian inference that are scalable to large inference problems.

DTSTART;TZID=America/New_York:20160128T133000 DTEND;TZID=America/New_York:20160128T143000 SEQUENCE:0 SUMMARY:Seminar: Tamara Broderick (MIT) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-tamara-broderick-mit-whi tehead-304/ END:VEVENT BEGIN:VEVENT UID:ai1ec-6197@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Title : On the spectra of direct sums and Kronecker products of side length 2 hypermatrices and related algorithmic problems in data s cience.

\nAbstract:

\nWe present elementary method for obtaini ng the spectral decomposition of hypermatrices generated by arbitrary comb inations of Kronecker products and direct sums of cubic hypermatrices havi ng side length 2. The method is based on a generalization of Parseval’s id entity. We use the general formulation of Parseval’s identity to introduc e hypermatrix Fourier transforms and discrete Fourier hypermatrices. We ex tend to hypermatrices orthogonalization procedures and Sylvester’s classic al Hadamard matrix construction. We conclude the talk with illustrations o f spectral decompositions of adjacency hypermatrices of finite groups and a proof of a hypermatrix Rayleigh quotient inequality.

\nThis is a j oint work with Yuval Filmus.

DTSTART;TZID=America/New_York:20160202T133000 DTEND;TZID=America/New_York:20160202T143000 SEQUENCE:0 SUMMARY:Seminar: Edinah Gnang (Purdue University) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-edinah-gnang-purdue-univ ersity-whitehead-304/ END:VEVENT BEGIN:VEVENT UID:ai1ec-6098@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Title: Using Integer Programming for Solving Nonconvex Quadr atic Programs with Box Constraints

\n\n

We discuss effective computational techniques for solving nonconvex quadratic programs with box constraints (BoxQP). Cutting planes obtained from the well-known Boolean Quadric Polytope may be applied in this context\, and we demonstrate the equivalence between the Chvatal-Gomory closure of a natural linear relaxat ion of (BoxQP) and the relaxation of the Boolean Quadric Polytope consisti ng of the odd-cycle inequalities. By using these cutting planes effective ly at nodes of the branch-and-bound tree\, in conjunction with additional integrality-based branching and a strengthened convex quadratic relaxation \, we demonstrate that we can effectively solve a well-known family of tes t instances. Our new solver\, GuBoLi\, is orders of magnitude faster than existing commercial and open-source solvers.

DTSTART;TZID=America/New_York:20160204T133000 DTEND;TZID=America/New_York:20160204T143000 SEQUENCE:0 SUMMARY:Seminar: Jeff Linderoth (University of Wisconsin- Madison) @ Whiteh ead 304 URL:https://engineering.jhu.edu/ams/events/seminar-jeff-linderoth-universit y-of-wisconsin-madison-whitehead-304/ END:VEVENT BEGIN:VEVENT UID:ai1ec-6185@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Title: Important Features PCA (IF-PCA) for Large-Scale Inf
erence\, with Applications in Gene Microarrays

\n\n\n**\n****\n**~~\n~~**\n****
DTSTART;TZID=America/New_York:20171006T100000
DTEND;TZID=America/New_York:20171006T110000
SEQUENCE:0
SUMMARY:Additional AMS Seminar: David Bergman (UConn) @ Whitehead 304
URL:https://engineering.jhu.edu/ams/events/additional-ams-seminar-david-ber
gman-uconn-whitehead-304/
X-COST-TYPE:free
END:VEVENT
BEGIN:VEVENT
UID:ai1ec-10669@engineering.jhu.edu/ams
DTSTAMP:20180317T041026Z
CATEGORIES:
CONTACT:
DESCRIPTION:****\n****\n****\n**

Abstract:

\nIdentification of sample labels is a majo
r problem in statistics with many applications. In the Big Data era\, it f
aces two main challenges: 1. the number of features is much larger than th
e sample size\; 2. the signals are sparse and weak\, masked by large amoun
t of noise.

\n\n\nWe propose a new tuning-
free clustering procedure for high-dimensional data\, Important Features P
CA (IF-PCA). IF-PCA consists of a feature selection step\, a PCA step\, an
d a k-means step. The first two steps reduce the data dimensions recursive
ly\, while the main information is preserved. As a consequence\, IF-PCA is
fast and accurate\, producing competitive performance in application to 1
0 gene microarray data sets.

\n\n\nWe also
generalize IF-PCA for the signal recovery and hypothesis testing problems
. With IF-PCA and two aggregation methods\, we find the statistical limits
for these three problems.

DTSTART;TZID=America/New_York:20160209T133000
DTEND;TZID=America/New_York:20160209T143000
SEQUENCE:0
SUMMARY:Seminar: Wanjie Wang (University of Pennsylvania) @ Whitehead 304
URL:https://engineering.jhu.edu/ams/events/seminar-wanjie-wang-university-o
f-pennsylvania/
END:VEVENT
BEGIN:VEVENT
UID:ai1ec-6344@engineering.jhu.edu/ams
DTSTAMP:20180317T041026Z
CATEGORIES:
CONTACT:
DESCRIPTION:Stochastic evolutionary modeling of cancer development and r esistance to treatment

\nCancer is the result of a stochastic evolut ionary process characterized by the accumulation of mutations that are res ponsible for tumor growth\, immune escape\, and drug resistance\, as well as mutations with no effect on the phenotype. Stochastic modeling can be u sed to describe the dynamics of tumor cell populations and obtain insights into the hidden evolutionary processes leading to cancer. I will present recent approaches that use branching process models of cancer evolution to quantify intra-tumor heterogeneity and the development of drug resistance \, and their implications for interpretation of cancer sequencing data and the design of optimal treatment strategies.

DTSTART;TZID=America/New_York:20160210T133000 DTEND;TZID=America/New_York:20160210T143000 SEQUENCE:0 SUMMARY:Seminar: Ivana Bozic (Harvard University) URL:https://engineering.jhu.edu/ams/events/seminar-ivana-bozic-harvard-univ ersity/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-6188@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Title: Universality in numerical computations with random da ta

\n\n

Abstract: This talk will concern recent progress on the statistical analysis of numerical algorithms with random initial data. In particular\, with appropriate randomness\, the fluctuations of the it eration count (halting time) of numerous numerical algorithms have been de monstrated to be universal\, i.e.\, independent of the distribution on the initial data. This phenomenon has given new insights into random matrix t heory. Furthermore\, estimates from random matrix theory allow for fluctua tion limit theorems for simple algorithms and halting time estimates for o thers. The universality in the halting time is directly related to the ex perimental work of Bakhtin and Correll on neural computation and human dec ision-making times.

DTSTART;TZID=America/New_York:20160211T133000 DTEND;TZID=America/New_York:20160211T143000 SEQUENCE:0 SUMMARY:Seminar: Tom Trogdon (New York University) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-tom-trogdon-new-york-uni versity-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-6348@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Bringing Moneyball to Campaigns

\n\n

Over the p ast decade\, an entire industry has grown up around the use of data to hel p campaigns be more efficient and effective. Whether it is trying to iden tify that last persuadable voter or allocating resources to get your suppo rters out to the polls\, today’s campaigns often rely on a staff of data a nalysts\, statisticians and modelers. Together\, data and analytics help identify which voters to target and what actions to take to generate the v otes where they are needed.

\nIn this talk I will introduce the tool s and techniques involving data\, analytics\, and experimentation used by campaigns. We will discuss where many of these techniques came from and h ow they evolved in politics to culminate in President Obama’s 2012 campaig n. This survey of the data-driven campaigns will include polling\, micro- targeting and random controlled experiments.

\n\n

Please join HUSAM in welcoming Dr. Ben Yuhas to the Johns Hopkins University communit y!

DTSTART;TZID=America/New_York:20160211T190000 DTEND;TZID=America/New_York:20160211T200000 SEQUENCE:0 SUMMARY:HUSAM: Dr. Ben Yuhas (Principal of the Yuhas Consulting Group\, LLC ) @ Gilman 50 URL:https://engineering.jhu.edu/ams/events/husam-dr-ben-yuhas-principal-of- the-yuhas-consulting-group-llc-gilman-50/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-6198@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Title: Controlling a Thermal Fluid: Theoretical and Computat ional Issues

\nAbstract: We first discuss the problem of designing a feedback law which locally stabilizes a two dimensional thermal fluid modeled by the Boussinesq equations. The problem was motivated by the des ign and operation of low energy consumption buildings. The investigation o f stability for a fluid flow in the natural convection problem is importan t in the theory of hydrodynamical stability. The challenge of stabilizati on of the Boussinesq equations arises from the stabilization of the Navier -Stokes equations and its coupling with the convection-diffusion equation for temperature. In our current work\, we are interested in stabilizing a possible unstable steady state solution to the Boussinesq equations on a b ounded and connected domain. We show that a finite number of controls acti ng on a part of the boundary through Neumann/Robin boundary conditions is sufficient to stabilize the full nonlinear equations in the neighborhood of this steady state solution. Dirichlet boundary conditions are imposed o n the rest of the boundary. Moreover\, we prove that a stabilizing feedbac k control law can be obtained based on the partial estimation of the sys tem state by solving an extended Kalman filter problem for the linearized Boussinesq equations. In particular\, a reduced order model is derived to construct a finite dimensional estimator. Numerical results are provid ed to illustrate the idea. In the end\, we discuss the problem of contr ol design for the Boussinesq equations with zero diffusivity and its appl ication to optimal mixing\, mass and energy transport during processing. DTSTART;TZID=America/New_York:20160216T133000 DTEND;TZID=America/New_York:20160216T143000 SEQUENCE:0 SUMMARY:Seminar: Weiwei Hu (University of Minnesota) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-weiwei-hu-university-of- minnesota-whitehead-304/ END:VEVENT BEGIN:VEVENT UID:ai1ec-6082@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:

Robust and efficient collocation methods for parameterized m odels

\nMonte Carlo (MC) methods for the construction of polynomial approximations are effective tools for building a computational surrogate of the parametric variation for a model response. In this talk we investig ate least-squares regularization of noisy data and compressive sampling re covery of sparse representations. We wish to minimize the number of sample s required for a stable and accurate procedure. We propose an algorithm fo r a particular kind of weighted Monte Carlo approximation method based on sampling from the pluripotential equilibrium measure. Standard MC methods suffer from poor stability and accuracy for high-order approximations\, bu t the properties of the equilibrium measure allow us to derive quasi-optim al statements of mathematical recoverability in both over- or undersampled regression problems. We also show that such an approach typically yields very stable\, high-order computational algorithms for parameterized PDE ap proximation. We present theoretical analysis to motivate the algorithm\, a nd numerical results to illustrate that equilibrium measure-based approach es are superior to standard MC methods in many situations of interest\, no tably in high-dimensional scenarios.

DTSTART;TZID=America/New_York:20160218T133000 DTEND;TZID=America/New_York:20160218T143000 SEQUENCE:0 SUMMARY:Seminar: Akil Narayan (University of Utah) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-akil-narayan-university- of-utah-whitehead-304/ END:VEVENT BEGIN:VEVENT UID:ai1ec-6199@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:**Title:**

Structure-Enhancing Algorith ms for Statistical Learning Problems

\n**Abstract:**

For many problems in statistical machine learning and data-driven de
cision-making\, massive datasets necessitate the use of scalable algorithm
s that deliver sensible (interpretable) and statistically sound solutions.
In this talk\, we discuss several scalable algorithms that directly prom
ote *well-structured *solutions in two related contexts: (i) sparse
high-dimensional linear regression\, and (ii) low-rank matrix completion\
, both of which are particularly relevant in modern machine learning. In
the context of linear regression\, we study several boosting algorithms –
which directly promote sparse solutions – from the perspective of modern f
irst-order methods in convex optimization. We use this perspective to der
ive the first-ever computational guarantees for existing boosting methods
and to develop new algorithms with associated computational guarantees as
well. In the context of matrix completion\, we present an extension of th
e Frank-Wolfe method in convex optimization that is designed to induce nea
r-optimal low-rank solutions for regularized matrix completion problems\,
and we derive computational guarantees that trade off between low-rank str
ucture and data fidelity. For both problem contexts\, we present computat
ional results using datasets from microarray and recommender system applic
ations.

Title: Recent theoretic and algorithmic advances in graph m atching

\nAbstract: Inference across multiple graphs arises natural ly in disciplines as varied as neuroscience\, physics\, and sociology. In a number of methodologies for joint inference across graphs\, however\, i t is assumed that an explicit vertex correspondence is a priori known acro ss the vertex sets of the graphs. While this assumption is often reasonabl e\, in practice these correspondences may be unobserved and/or errorfully observed\, and graph matching—aligning a pair of graphs to minimize their edge disagreements—is used to align the graphs before performing subsequen t inference. Graph matching is a computationally challenging and well-stu died problem\, but few existing algorithms have theoretical support for th eir performance. For tractability\, many algorithms begin by relaxing the problem’s binary constraints\, thus rendering applicable gradient-descent methodologies. We develop a state-of-the-art algorithm for solving an ind efinite relaxed graph matching problem\, and we show that under mild model assumptions\, our indefinite relaxation (when solved exactly) almost alwa ys uncovers the optimal permutation\, while the commonly used convex relax ation almost always fails to identify the optimal permutation. We highlig ht some of the practical and theoretical implications of these results on real and synthetic data\, and we discuss recent work towards formalizing t he connection between graph matching and pairwise mutual information.

DTSTART;TZID=America/New_York:20160225T133000 DTEND;TZID=America/New_York:20160225T143000 SEQUENCE:0 SUMMARY:Seminar: Vince Lyzinski (JHU) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-vince-lyzinski-jhu/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-6111@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Mediation: From Intuition to Data Analysis.

\n\n< p>Modern causal inference links the “top-down” representation of causal in tuitions and “bottom-up” data analysis with the aim of choosing policy. Tw o innovations that proved key for this synthesis were a formalization of H ume’s counterfactual account of causation using potential outcomes (due to Jerzy Neyman)\, and viewing cause effect relationships via directed acycl ic graphs (due to Sewall Wright). I will briefly review how a synthesis o f these two ideas was instrumental in formally representing the notion of “causal effect” as a parameter in the language of potential outcomes\, and discuss a complete identification theory linking these types of causal pa rameters and observed data\, as well as approaches to estimation of the re sulting statistical parameters.\n

I will then describe\, in more det ail\, how my collaborators and I are applying the same approach to mediati on\, the study of effects along particular causal pathways. I consider me diated effects at their most general: I allow arbitrary models\, the prese nce of hidden variables\, multiple outcomes\, longitudinal treatments\, an d effects along arbitrary sets of causal pathways. As was the case with c ausal effects\, there are three distinct but related problems to solve — a representation problem (what sort of potential outcome does an effect alo ng a set of pathways correspond to)\, an identification problem (can a cau sal parameter of interest be expressed as a functional of observed data)\, and an estimation problem (what are good ways of estimating the resulting statistical parameter). I report a complete solution to the first two pr oblems\, and progress on the third. In particular\, my collaborators and I show that for some parameters that arise in mediation settings\, triply robust estimators exist\, which rely on an outcome model\, a mediator mode l\, and a treatment model\, and which remain consistent if any two of thes e three models are correct.

\n\n

Some of the reported results are a joint work with Eric Tchetgen\, Caleb Miles\, Phyllis Kanki\, and S eema Meloni.

DTSTART;TZID=America/New_York:20160303T133000 DTEND;TZID=America/New_York:20160303T143000 SEQUENCE:0 SUMMARY:Seminar: Ilya Shpitser (Johns Hopkins University) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-ilya-shpitzer-johns-hopk ins-university-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-6462@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Title: Arbitrage-Free Pricing of XVA.

\n\n

Abst ract: We develop a framework for computing the total valuation adjustment (XVA) of a European claim accounting for funding costs\, counterparty cred it risk\, and collateralization. Based on no-arbitrage arguments\, we deri ve backward stochastic differential equations

\n(BSDEs) associated w ith the replicating portfolios of long and short positions in the claim. T his leads to the definition of buyer’s and seller’s XVA\, which in turn id entify a no-arbitrage interval. In the case that borrowing and lending rat es coincide\, we provide a fully explicit expression for the uniquely dete rmined XVA\, expressed as a percentage of the price of the traded claim\, and for the corresponding replication strategies. In the general case of a symmetric funding\, repo and collateral rates\, we study the semi-linear p artial differential equation (PDE) characterizing buyer’s and seller’s XVA and show the existence of a unique classical solution to it. To illustrat e our results\, we conduct a numerical study demonstrating how funding cos ts\, repo rates\, and counterparty risk contribute to determine the total valuation adjustment. This talk is based on joint works with Agostino Capp oni (Columbia) and Stephan Sturm (WPI).

DTSTART;TZID=America/New_York:20160324T133000 DTEND;TZID=America/New_York:20160324T143000 SEQUENCE:0 SUMMARY:Seminar: Maxim Bichuch (JHU) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-maxim-bichuch-jhu-whiteh ead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-6116@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Title: Co-clustering of nonsmooth graphons

\nAbstract:

\nTheoretical results are becomming known for community detection a nd clustering of networks\; however\, these results assume an idealized ge nerative model that is unlikely to hold in many settings. Here we consider exploratory co-clustering of a bipartite network\, where the rows and col umns of the adjacency matrix are assumed to be samples from an arbitrary p opulation. This is equivalent to assuming that the data is generated from a nonparametric model known as a graphon. We show that co-clusters found b y any method can be extended to the row and column populations\, or equiva lently that the estimated blockmodel approximates a blocked version of the generative graphon\, with generalization error bounded by n^{-1/2}. Analo gous results are also shown for degree-corrected co-blockmodels and random dot product bipartite graphs\, with error rates depending on the dimensio nality of the latent variable space.

DTSTART;TZID=America/New_York:20160331T133000 DTEND;TZID=America/New_York:20160331T143000 SEQUENCE:0 SUMMARY:Seminar: David Choi (CMU) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-david-choi-cmu-whitehead -304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-6466@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:**Title: **Distributed proximal gradient method
s for cooperative multi-agent consensus optimization

** **

**Abstract:**

In this talk\, I will discuss decentralized methods for solving cooperative multi -agent consensus optimization problems. Consider an undirected network of agents\, where only those agents connected by an edge can directly communi cate with each other. The objective is to minimize the sum of agent-specif ic composite convex functions\, i.e.\, each term in the sum is a private c ost function belonging to an agent. In the first part\, I will discuss the unconstrained case\, and in the second part I will focus on the constrain ed case\, where each agent has a private conic constraint set. For the con strained case the optimal consensus decision should lie in the intersectio n of these private sets. This optimization model abstracts a number of app lications in machine learning\, distributed control\, and estimation using sensor networks. I will discuss different types of distributed algorithms \; in particular\, I will describe methods based on inexact augmented Lagr angian\, and linearized ADMM. I will provide convergence rates both in sub -optimality error and consensus violation\; and also examine the effect of underlying network topology on the convergence rates of the proposed dece ntralized algorithms.

\n\n

Joint work with Ph.D. students Zi Wang\, Erfan Yazdandoost\, and Shiqian Ma from Chinese University of Hong Kong\, and Garud Iyengar from Columbia University.

DTSTART;TZID=America/New_York:20160407T133000 DTEND;TZID=America/New_York:20160407T143000 SEQUENCE:0 SUMMARY:Seminar: Serhat Aybat (Penn State University) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-serhat-aybat-penn-state- university-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-6084@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Title: Scalable Bayesian Models of Interacting Time Series\n

\n

Abstract:

\nData streams of increasing complexity a nd scale are being collected in a variety of fields ranging from neuroscie nce\, genomics\, and environmental monitoring to e-commerce. Modeling the intricate and possibly evolving relationships between the large collectio n of series can lead to increased predictive performance and domain-interp retable structures. For scalability\, it is crucial to discover and explo it sparse dependencies between the data streams. Such representational st ructures for independent data sources have been studied extensively\, but have received limited attention in the context of time series. In this ta lk\, we present a series of Bayesian models for capturing such sparse depe ndencies via clustering\, graphical models\, and low-dimensional embedding s of time series. We explore these methods in a variety of applications\ , including house price modeling and inferring networks in the brain.

\nWe then turn to observed interaction data\, and briefly touch upon ho w to devise statistical network models that capture important network feat ures like sparsity of edge connectivity. Within our Bayesian framework\, a key insight is to move to a continuous-space representation of the graph \, rather than the typical discrete adjacency matrix structure. We demons trate our methods on a series of real-world networks with up to hundreds o f thousands of nodes and millions of edges.

\n\n

Bio:

\nEmily Fox is currently the Amazon Professor of Machine Learning in the St atistics Department at the University of Washington. She received a S.B. in 2004 and Ph.D. in 2009 from the Department of Electrical Engineering an d Computer Science at MIT. She has been awarded a Sloan Research Fellowsh ip (2015)\, an ONR Young Investigator award (2015)\, an NSF CAREER award ( 2014)\, the Leonard J. Savage Thesis Award in Applied Methodology (2009)\, and the MIT EECS Jin-Au Kong Outstanding Doctoral Thesis Prize (2009). He r research interests are in large-scale Bayesian dynamic modeling and comp utations.

DTSTART;TZID=America/New_York:20160414T133000 DTEND;TZID=America/New_York:20160414T143000 SEQUENCE:0 SUMMARY:Seminar: Emily Fox (University of Washington) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-emily-fox-university-of- washington-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-6032@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Title: Quickest detection in correlated and coupled systems< /p>\n

Abstract:

\nIn this works we consider the problem N-dimension
al quickest detection in correlated and coupled systems. The objective is
to detect the first time that the system of N sensors undergoes a change w
ith a one shot communication to the central fusion center.

\nIn both
cases it is seen that the minimum of N – cumulative sum tests with appropr
iately chosen thresholds is asymptotically optimal in managing the trade o
ff between a small detection delay and a large mean time to first False al
arm as the mean time to the first false alarm increases without bound. In
the former case a Linear penalty is used for detection delay while in the
latter a Kulback- Leibler distance of the measure before and after regime
switching is used.

**Movie Reconstruction from Brain Signals: “Mind-Readi
ng”**

In a thrilling breakthrough at the intersection of ne uroscience and statistics\, penalized Least Squares methods have been used to construct a “mind-reading” algorithm that reconstructs movies from fMR I brain signals. The story of this algorithm is a fascinating tale of the interdisciplinary research that led to the development of the system which was selected as one of Time Magazine’s 50 Best Inventions of 2011. Talk 1 : Movie Reconstruction from Brain Signals: “Mind-Reading”

DTSTART;TZID=America/New_York:20160427T133000 DTEND;TZID=America/New_York:20160427T143000 SEQUENCE:0 SUMMARY:Duncan Lecture Series: Bin Yu (University of California Berkeley) @ Gilman 50 URL:https://engineering.jhu.edu/ams/events/duncan-lecture-series-bin-yu-uni versity-of-california-berkeley-2/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-6112@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:**Unveiling the mysteries in spatial gene expression**

Genome-wide data reveal an intricate landscape where gene activities are highly differentiated across diverse spatial areas. These g ene actions and interactions play a critical role in the development and f unction of both normal and abnormal tissues. As a result\, understanding s patial heterogeneity of gene networks is key to developing treatments for human diseases. Despite the abundance of recent spatial gene expression da ta\, extracting meaningful information remains a challenge for local gene interaction discoveries. In response\, we have developed staNMF\, a method that combines a powerful unsupervised learning algorithm\, nonnegative ma trix factorization (NMF)\, with a new stability criterion that selects the size of the dictionary. Using staNMF\, we generate biologically meaningfu l Principle Patterns (PP)\, which provide a novel and concise representati on of Drosophila embryonic spatial expression patterns that correspond to pre-organ areas of the developing embryo. Furthermore\, we show how this n ew representation can be used to automatically predict manual annotations\ , categorize gene expression patterns\, and reconstruct the local gap gene network with high accuracy. Finally\, we discuss on-going crispr/cas9 kno ck-out experiments on Drosophila to verify predicted local gene-gene inter actions involving gap-genes. An open-source software is also being built b ased on SPARK and Fiji.

\nThis talk is based on collaborative work o f a multi-disciplinary team (co-lead Erwin Frise) from the Yu group (stati stics) at UC Berkeley\, the Celniker group (biology) at the Lawrence Berke ley National Lab (LBNL)\, and the Xu group (computer science) at Hsinghua Univ.

DTSTART;TZID=America/New_York:20160428T133000 DTEND;TZID=America/New_York:20160428T143000 SEQUENCE:0 SUMMARY:Duncan Lecture Series: Bin Yu (University of California Berkeley) @ Mergenthaler 111 URL:https://engineering.jhu.edu/ams/events/duncan-lecture-series-bin-yu-uni versity-of-california-berkeley/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-6970@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION: DTSTART;TZID=America/New_York:20160901T133000 DTEND;TZID=America/New_York:20160901T143000 SEQUENCE:0 SUMMARY:Seminar: Getting to Know You @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-getting-to-know-you-whit ehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-7268@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT:http://www.math.jhu.edu/~data/ DESCRIPTION:Statistics of the Stability Bounds in the Phase Retrieval Pr
oblem

\nIn this talk we present a local-global Lipschitz analysis of
the phase retrieval problem. Additionally we present tentative estimates o
f the tail-bound for the distribution of the global Lipschitz constants. S
pecifically it is known that if the frame {f1\,…\,fm} for Cn is phase retr
ievable then there are constants a0 and b0 so that for every x\,y∈Cn: a0∣∣
xx*-yy*∣∣12≤∑k=1m∣∣〈x\,fk〉∣2-∣〈y\,fk〉∣2∣2≤b0∣∣xx*-yy*∣∣12. Assumef1\,…\,fm
are independent realizations with entries from CN(0\,1). In this talk we
establish estimates for the probability P(a0>a).

Title: To Replace or Not to Replace in Finite Population Sam pling

\nAbstract:

\nWe revisit the classical result in finite
population sampling which states that in *equally-likely *“simple”
random sampling the sample mean is more reliable when we do not replace af
ter each draw. In this talk\, we review a classical result for the equall
y likely sampling case. Then we investigate if and when the same is true f
or samples where it may no longer be true that each member of the populati
on has an equal chance of being selected\, and when the population mean is
estimated using the Horvitz-Thompson inverse probability weighing to prod
uce an unbiased estimator. For a certain class of sampling schemes\, we a
re able to obtain convenient expressions for the variance of the sample me
an and surprisingly\, we find that for some selection distributions a more
reliable estimate of the population mean will happen by replacing after e
ach draw. We show for selection distributions lying in a certain polytope
the classical result prevails.

\n

This is joint work with Fred Torcaso.

DTSTART;TZID=America/New_York:20160908T133000 DTEND;TZID=America/New_York:20160908T143000 SEQUENCE:0 SUMMARY:Seminar: Dan Naiman (JHU) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-dan-naiman-jhu-whitehead -304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-7272@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT:http://www.math.jhu.edu/~data/ DESCRIPTION:Finite-Sample Bounds for Geometric Multires
olution Analysis

\nDTSTART;TZID=America/New_York:20160914T150000 DTEND;TZID=America/New_York:20160914T160000 SEQUENCE:0 SUMMARY:Data Seminar: Nate Strawn (Georgetown University) @ Krieger 309 URL:https://engineering.jhu.edu/ams/events/data-seminar-nate-strawn-georget own-university-krieger-309/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-6996@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:

**S
tochastic Newton Methods for Machine Learning**

\n

Jorge Nocedal is the David and Karen Sachs Professor of Industr ial Engineering and Management Sciences at Northwestern University. He rec eived his PhD in Mathematical Sciences from Rice University and was a post doctoral fellow at the Courant Institute. His research is in nonlinear opt imization with applications to machine learning. Over the years\, his work has spanned algorithms\, analysis and software. He is a SIAM Fellow\, has been an invited speaker at the International Congress of Mathematicians\, and was awarded the 2012 George B. Dantzig Prize.

\nDTSTART;TZID=America/New_York:20160915T133000 DTEND;TZID=America/New_York:20160915T143000 SEQUENCE:0 SUMMARY:Goldman Lecture Series: Jorge Nocedal (Northwestern University)- Ma ryland 110 URL:https://engineering.jhu.edu/ams/events/goldman-lecture-series-jorge-noc edal-northwestern-university-location-tbd/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-7352@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:

The JHU Actuarial Club will host an event on September 16th. The speaker JHU alum\, Matt Sedlock\, is currently working at Mass Mutua l. He will share his experience working in the actuarial industry and dis cuss the recruiting process from Mass Mutual

DTSTART;TZID=America/New_York:20160916T180000 DTEND;TZID=America/New_York:20160916T200000 SEQUENCE:0 SUMMARY:What is an Actuary? @ Arellano Theatre\, Levering Hall URL:https://engineering.jhu.edu/ams/events/actuary-arellano-theatre-leverin g-hall/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-7276@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT:http://www.math.jhu.edu/~data/ DESCRIPTION: DTSTART;TZID=America/New_York:20160921T150000 DTEND;TZID=America/New_York:20160921T160000 SEQUENCE:0 SUMMARY:Data Seminar: Charles Meneveau (JHU) @ Krieger 309 URL:https://engineering.jhu.edu/ams/events/data-seminar-charles-meneveau-jh u-shaffer-100/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-6964@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:**High-Dimensional Analysis of Stochastic Algorithms f
or Convex and Nonconvex Optimization: Limiting Dynamics and Phase Transiti
ons**

\n

**Abstract**

We consid er efficient iterative methods (e.g.\, stochastic gradient descent\, rando mized Kaczmarz algorithms\, iterative coordinate descent) for solving larg e-scale optimization problems\, whether convex or nonconvex. A flurry of r ecent work has focused on establishing their theoretical performance guara ntees. This intense interest is spurred on by the remarkably impressive em pirical performance achieved by these low-complexity and memory-efficient methods.

\nIn this talk\, we will present a framework for analyzing the exact dynamics of these methods in the high-dimensional limit. For con creteness\, we consider two prototypical problems: regularized linear regr ession (e.g. LASSO) and sparse principal component analysis. For each case \, we show that the time-varying estimates given by the algorithms will co nverge weakly to a deterministic “limiting process” in the high-dimensiona l (scaling and mean-field) limit. Moreover\, this limiting process can be characterized as the unique solution of a nonlinear PDE\, and it provides exact information regarding the asymptotic performance of the algorithms. For example\, performance metrics such as the MSE\, the cosine similarity and the misclassification rate in sparse support recovery can all be obtai ned by examining the deterministic limiting process. A steady-state analys is of the nonlinear PDE also reveals interesting phase transition phenomen ons related to the performance of the algorithms. Although our analysis is asymptotic in nature\, numerical simulations show that the theoretical pr edictions are accurate for moderate signal dimensions.

\nWhat makes our analysis tractable is the notion of exchangeability\, a fundamental pr operty of symmetry that is inherent in many of the optimization problems e ncountered in signal processing and machine learning.

\n\n

~~Bio~~

Yue M. Lu was born in Shanghai. After finishing undergraduate studies at Shanghai Jiao Tong University\, he attended the University of Illinois at Urbana-Champaign\, where he received the M.Sc. d egree in mathematics and the Ph.D. degree in electrical engineering\, both in 2007. He was a Research Assistant at the University of Illinois at Urb ana-Champaign\, and has worked for Microsoft Research Asia\, Beijing\, and Siemens Corporate Research\, Princeton\, NJ. Following his work as a post doctoral researcher at the Audiovisual Communications Laboratory at Ecole Polytechnique Fédérale de Lausanne (EPFL)\, Switzerland\, he joined Harvar d University in 2010\, where he is currently an Associate Professor of Ele ctrical Engineering at the John A. Paulson School of Engineering and Appli ed Sciences.

\nHe received the Most Innovative Paper Award (with Min h N. Do) of IEEE International Conference on Image Processing (ICIP) in 20 06\, the Best Student Paper Award of IEEE ICIP in 2007\, and the Best Stud ent Presentation Award at the 31st SIAM SEAS Conference in 2007. Student p apers supervised and coauthored by him won the Best Student Paper Award (w ith Ivan Dokmanic and Martin Vetterli) of IEEE International Conference on Acoustics\, Speech and Signal Processing in 2011 and the Best Student Pap er Award (with Ameya Agaskar and Chuang Wang) of IEEE Global Conference on Signal and Information Processing (GlobalSIP) in 2014.

\nHe has bee n an Associate Editor of the IEEE Transactions on Image Processing since 2 014\, an Elected Member of the IEEE Image\, Video\, and Multidimensional S ignal Processing Technical Committee since 2015\, and an Elected Member of the IEEE Signal Processing Theory and Methods Technical Committee since 2 016. He received the ECE Illinois Young Alumni Achievement Award in 2015.< /p> DTSTART;TZID=America/New_York:20160922T133000 DTEND;TZID=America/New_York:20160922T143000 SEQUENCE:0 SUMMARY:Seminar: Yue Lu (Harvard) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-yue-lu-harvard-whitehead -304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-7280@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT:http://www.math.jhu.edu/~data/ DESCRIPTION:

From Molecular Dynamics to Large Scale Infe
rence

\n

\nMolecular models and data analyt
ics problems give rise to very large systems of stochastic differential eq
uations (SDEs) whose paths are designed to ergodically sample multimodal p
robability distributions. An important challenge for the numerical analyst
(or the data scientist\, for that matter) is the design of numerical proc
edures to generate these paths. One of the interesting ideas is to constru
ct stochastic numerical methods with close attention to the error in the i
nvariant measure. Another is to redesign the underlying stochastic dynamic
s to reduce bias or locally transform variables to enhance sampling effici
ency. I will illustrate these ideas with various examples\, including a ge
odesic integrator for constrained Langevin dynamics [1] and an ensemble sa
mpling strategy for distributed inference [2].

Title: Edge-coloring Multigraphs

\nAbstract: Graph (ve rtex) coloring is a central area of discrete math\; however\, it is NP-har d even to approximate the chromatic number. Edge-coloring can be seen as a special case of vertex coloring. As such\, we may hope that computing ( or approximating) the edge chromatic number is easier\; in fact\, it is. We will survey a number of theorems and conjectures on edge-coloring. The se include many instances when the edge-chromatic number satisfies a trivi al lower bound with equality\, such as when it equals the graph’s maximum degree. I will also mention some of my recent work in this area\, and int roduce one of the main tools\, Tashkinov trees\, which rely on a beautiful double induction.

DTSTART;TZID=America/New_York:20160929T133000 DTEND;TZID=America/New_York:20160929T143000 SEQUENCE:0 SUMMARY:Seminar: Dan Cranston (Virginia Commonwealth University) @ Whitehea d 304 URL:https://engineering.jhu.edu/ams/events/seminar-dan-cranston-virginia-co mmonwealth-university-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-7439@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT:http://www.math.jhu.edu/~data/ DESCRIPTION:Geometric Methods for the Approximation of
High-dimensional Dynamical Systems

\n

\nI w
ill discuss a geometry-based statistical learning framework for performing
model reduction and modeling of stochastic high-dimensional dynamical sys
tems. I will consider two complementary settings. In the first one\, I am
given long trajectories of a system\, e.g. from molecular dynamics\, and I
discuss techniques for estimating\, in a robust fashion\, an effective nu
mber of degrees of freedom of the system\, which may vary in the state spa
ce of then system\, and a local scale where the dynamics is well-approxima
ted by a reduced dynamics with a small number of degrees of freedom. I wil
l then use these ideas to produce an approximation to the generator of the
system and obtain\, via eigenfunctions of an empirical Fokker-Planck ques
tion\, reaction coordinates for the system that capture the large time beh
avior of the dynamics. I will present various examples from molecular dyna
mics illustrating these ideas. In the second setting I assume I only have
access to a (large number of expensive) simulators that can return short s
imulations of high-dimensional stochastic system\, and introduce a novel s
tatistical learning framework for learning automatically a family of local
approximations to the system\, that can be (automatically) pieced togethe
r to form a fast global reduced model for the system\, called ATLAS. ATLAS
is guaranteed to be accurate (in the sense of producing stochastic paths
whose distribution is close to that of paths generated by the original sys
tem) not only at small time scales\, but also at large time scales\, under
suitable assumptions on the dynamics. I discuss applications to homogeniz
ation of rough diffusions in low and high dimensions\, as well as relative
ly simple systems with separations of time scales\, and deterministic chao
tic systems in high-dimensions\, that are well-approximated by stochastic
differential equations.

*No knowledge of molecular dynamics is r
equired\, and the techniques above are quite universal. Ideas in the first
part of the talk are based on what is called Diffusion Geometry\, and hav
e been used widely in data analysis\; ideas in the second part are applica
ble to MCMC. The talk will be accessible to students with a wide variety o
f backgrounds and interests.*

Title:

\nStochastic Search Methods for Simulation Opti mization

\n\n

Abstract:

\nA variety of systems arising in finance\, engineering design\, and manufacturing require the use of opt imization techniques to improve their performance. Due to the complexity a nd stochastic dynamics of such systems\, their performance evaluation freq uently requires computer simulation\, which however often lacks structure needed by classical optimization methods. We developed a gradient-based st ochastic search approach\, based on the idea of converting the original (s tructure-lacking) problem to a differentiable optimization problem on the parameter space of a sampling distribution that guides the search. A two-t imescale updating scheme is further studied and incorporated to improve th e algorithm efficiency. Convergence properties of our approach are establi shed through techniques from stochastic approximation\, and the performanc e of our algorithms is illustrated in comparison with some state-of-the-ar t simulation optimization methods. This is a joint work with Jiaqiao Hu (S tony Brook University) and Shalabh Bhartnagar (Indian Institute of Science ).

\n\n

Biography:

\nEnlu Zhou is currently an associat e professor in the H. Milton School of Industrial & Systems Engineering at Georgia Institute of Technology. Prior to joining Georgia Tech in 2013\, she was an assistant professor in the Industrial & Enterprise Systems Eng ineering Department at the University of Illinois Urbana-Champaign from 20 09-2013. She received the B.S. degree with highest honors in electrical en gineering from Zhejiang University\, China\, in 2004\, and the Ph.D. degre e in electrical engineering from the University of Maryland\, College Park \, in 2009. Her research interests include stochastic control\, simulation optimization\, and Monte Carlo statistical methods. She is a recipient of the “Best Theoretical Paper” award at the Winter Simulation Conference in 2009\, AFOSR Young Investigator award in 2012\, and NSF CAREER award in 2 015.

DTSTART;TZID=America/New_York:20161006T133000 DTEND;TZID=America/New_York:20161006T143000 SEQUENCE:0 SUMMARY:Seminar: Enlu Zhou (Georgia Tech) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-enlu-zhou-georgia-tech-w hitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-7463@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION: DTSTART;TZID=America/New_York:20161010T190000 DTEND;TZID=America/New_York:20161010T213000 SEQUENCE:0 SUMMARY:HUSAM Event: Global Hedge Fund Marshall Wace Recruiting Event URL:https://engineering.jhu.edu/ams/events/husam-event-global-hedge-fund-ma rshall-wace-recruiting-event/ X-COST-TYPE:free X-WP-IMAGES-URL:thumbnail\;http://engineering.jhu.edu/ams/wp-content/upload s/sites/44/2016/10/Husam-300x232.jpg\;730\;565\,medium\;http://engineering .jhu.edu/ams/wp-content/uploads/sites/44/2016/10/Husam-300x232.jpg\;730\;5 65\,large\;http://engineering.jhu.edu/ams/wp-content/uploads/sites/44/2016 /10/Husam-300x232.jpg\;730\;565\,full\;http://engineering.jhu.edu/ams/wp-c ontent/uploads/sites/44/2016/10/Husam-300x232.jpg\;730\;565 END:VEVENT BEGIN:VEVENT UID:ai1ec-7296@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT:http://www.math.jhu.edu/~data/ DESCRIPTION:Modeling the dynamics of interacting partic
les by means of stochastic networks

\n

\nMa
terial science have been rapidly developing in recent years. A variety of
particles interacting according to different kinds of pair potentials has
been produced in experimental works. Looking into the future\, one can ima
gine controlled self-assembly of particles into clusters of desired struct
ures leading to the creation of new types of materials. Analytical studies
of the self-assembly involve coping with difficulties associated with the
huge numbers configurations\, high dimensionality\, complex geometry\, an
d unacceptably large CPU times. A feasible approach to the study of self-a
ssembly consists of mapping the collections of clusters onto stochastic ne
tworks (continuous-time Markov chains) and analyzing their dynamics. Verti
ces of the networks represent local minima of the potential energy of the
clusters\, while arcs connect only those pairs of vertices that correspond
to local minima between which direct transitions are physically possible.
Transition rates along the arcs are the transition rates between the corr
esponding pairs of local minima. Such networks are mathematically tractabl
e and\, at the same time\, preserve important features of the underlying d
ynamics. Nevertheless\, their huge size and complexity render their analys
is challenging and invoke the development of new mathematical techniques.
I will discuss some approaches to construction and analysis of such networ
ks.

Title:

\nLeveraged Funds: Robust Replication and Perfo rmance Evaluation

\nAbstract:

\nLeveraged and inverse ETFs see k a daily return equal to a multiple of an index’ return\, an objective th at requires continuous portfolio rebalancing. The resulting trading costs create a tradeoff between tracking error\, which controls the short-term c orrelation with the index\, and excess return (or tracking difference) – t he long-term deviation from the leveraged index’ performance. With proport ional trading costs\, the optimal replication policy is robust to the inde x’ dynamics. A summary of a fund’s performance is the implied spread\, equ al to the product of tracking error and excess return\, rescaled for lever age and average volatility. The implied spread is insensitive to the bench mark’s risk premium and offers a tool to compare the performance of funds tracking the same index with different factors and tracking errors.

\n< p> \nhttp://ssrn.com/abstract=2839852

DTSTART;TZID=America/New_York:20161013T133000 DTEND;TZID=America/New_York:20161013T143000 SEQUENCE:0 SUMMARY:Seminar: Paolo Guasoni (Boston University) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-paolo-guasoni-boston-uni versity-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-7300@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT:http://www.math.jhu.edu/~data/ DESCRIPTION: DTSTART;TZID=America/New_York:20161026T150000 DTEND;TZID=America/New_York:20161026T160000 SEQUENCE:0 SUMMARY:Data Seminar: Robert Pego (Carnegie Mellon University) @ Krieger 30 9 URL:https://engineering.jhu.edu/ams/events/data-seminar-robert-pego-carnegi e-mellon-university-krieger-309/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-6963@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:**Adaptive Contrast Weighted Learning and Tree-based R
einforcement Learning for Multi-Stage Multi-Treatment Decision-Making**

Dynamic treatment regimes (DTRs) are sequential decision rule s that focus simultaneously on treatment individualization and adaptation over time. We develop robust and flexible semiparametric and machine learn ing methods for estimating optimal DTRs. In this talk\, we present a dynam ic statistical learning method\, adaptive contrast weighted learning (ACWL )\, which combines doubly robust semiparametric regression estimators with flexible machine learning methods. ACWL can handle multiple treatments at each stage and does not require prespecifying candidate DTRs. At each sta ge\, we develop robust semiparametric regression-based contrasts with the adaptation of treatment effect ordering for each patient\, and the adaptiv e contrasts simplify the problem of optimization with multiple treatment c omparisons to a weighted classification problem that can be solved with ex isting machine learning techniques. We further develop a tree-based reinfo rcement learning (T-RL) method to directly estimate optimal DTRs in a mult i-stage multi-treatment setting. At each stage\, T-RL builds an unsupervis ed decision tree that maintains the nature of batch-mode reinforcement lea rning. Unlike ACWL\, T-RL handles the optimization problem with multiple t reatment comparisons directly through the purity measure constructed with augmented inverse probability weighted estimators. By combining robust sem iparametric regression with flexible tree-based learning\, T-RL is robust\ , efficient and easy to interpret for the identification of optimal DTRs. However\, ACWL seems more robust to tree-type misspecification than T-RL w hen the true optimal DTR is non-tree-type. We illustrate the performances of both methods in simulations and case studies.

DTSTART;TZID=America/New_York:20161027T133000 DTEND;TZID=America/New_York:20161027T143000 SEQUENCE:0 SUMMARY:Seminar: Lu Wang (University of Michigan) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-lu-wang-university-of-mi chigan-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-7304@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT:http://www.math.jhu.edu/~data/ DESCRIPTION:Variational problems on graphs and their co
ntinuum limits

\n

\nWe will discuss variati
onal problems arising in machine learning and their limits as the number o
f data points goes to infinity. Consider point clouds obtained as random s
amples of an underlying “ground-truth” measure. Graph representing the poi
nt cloud is obtained by assigning weights to edges based on the distance b
etween the points. Many machine learning tasks\, such as clustering and cl
assification\, can be posed as minimizing functionals on such graphs. We c
onsider functionals involving graph cuts and graph laplacians and their li
mits as the number of data points goes to infinity. In particular we estab
lish under what conditions the minimizers of discrete problems have a well
defined continuum limit\, and characterize the limit. The talk is primari
ly based on joint work with Nicolas Garcia Trillos\, as well as on works w
ith Xavier Bresson\, Moritz Gerlach\, Matthias Hein\, Thomas Laurent\, Jam
es von Brecht and Matt Thorpe.

An Introduction to Distance Preserving Projections of Smooth Manifolds

\n\n

Manifold-based image models are assumed in ma ny engineering applications involving imaging and image classification. I n the setting of image classification\, in particular\, proposed designs f or small and cheap cameras motivate compressive imaging applications invol ving manifolds. Interesting mathematics results when one considers that t he problem one needs to solve in this setting ultimately involves question s concerning how well one can embed a low-dimensional smooth sub-manifold of high-dimensional Euclidean space into a much lower dimensional space wi thout knowing any of its detailed structure. We will motivate this proble m and discuss how one might accomplish this seemingly difficult task using random projections. Little if any prerequisites will be assumed beyond l inear algebra and some probability.

DTSTART;TZID=America/New_York:20161103T133000 DTEND;TZID=America/New_York:20161103T143000 SEQUENCE:0 SUMMARY:Seminar: Mark Iwen (Michigan State University) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-mark-iwen-michigan-state -university-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-7308@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT:http://www.math.jhu.edu/~data/ DESCRIPTION:Scalable Information Inequalities for Uncer
tainty Quantification in high dimensional probabilistic models

\n<
big>

\nIn this this talk we discuss new scalable informa
tion bounds for quantities of interest of complex stochastic models. The s
calability of inequalities allows us to (a) obtain uncertainty quantificat
ion bounds for quantities of interest in high-dimensional systems and/or f
or long time stochastic dynamics\; (b) assess the impact of large model pe
rturbations such as in nonlinear response regimes in statistical mechanics
\; (c) address model-form uncertainty\, i.e. compare different extended pr
obabilistic models and corresponding quantities of interest. We demonstrat
e these tools in fast sensitivity screening of chemical reaction networks
with a very large number of parameters\, and towards obtaining robust and
tight uncertainty quantification bounds for phase diagrams in statistical
mechanics models.

**Slipping Through the Cracks: Detecting Manipulation
in Regional Commodity Markets**

\n

Reid B. Stevens[1] and Jeffery Y. Zhang[2]

\n\n

Between 2010 and 2014\, the re
gional price of aluminum in the United States (Midwest premium) increased
400 percent. We argue that the Midwest premium was likely manipulated duri
ng this period through the exercise of market power in the aluminum storag
e market. We first use a difference-in-differences model to show that ther
e was a statistically significant increase of $0.07 per pound in the regio
nal price of aluminum relative to the regional price of a production compl
ement\, copper. We then use several instrumental variables to show that t
his increase was driven by a single financial company’s accumulation of an
unprecedented level of aluminum inventories in Detroit. Since this schem
e targeted the regional price of aluminum\, regulators who monitored only
spot and futures prices would not have noticed anything peculiar. We there
fore present an algorithm for real-time detection of similar manipulation
schemes in regional commodity markets. The algorithm confirms the existen
ce of a structural break in the U.S. aluminum market in late 2011. Using t
he algorithm\, regulators could have detected the scheme as early as Decem
ber 2012\, more than six months before it was publicized by an article in
*The New York Times*. We also apply the algorithm to another suspec
ted case of regional price manipulation in the European aluminum market an
d find a similar break in 2011\, suggesting the scheme may have been imple
mented beyond the United States.

[1] Department of Agriculture Economics\, Texas A&M University\, steve ns@tamu.edu

\n[2] Department of Economics\, Yale University and Harvard Law School\, jeffery.zhang@yale.e du

DTSTART;TZID=America/New_York:20161117T133000 DTEND;TZID=America/New_York:20161117T143000 SEQUENCE:0 SUMMARY:Seminar: Reid Stevens (Texas A&M University) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-helyette-geman-jhu-white head-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-7312@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT:http://www.math.jhu.edu/~data/ DESCRIPTION: DTSTART;TZID=America/New_York:20161130T150000 DTEND;TZID=America/New_York:20161130T160000 SEQUENCE:0 SUMMARY:Data Seminar: Youssef Marzouk (Massachusetts Institute of Technolog y) @ Krieger 309 URL:https://engineering.jhu.edu/ams/events/data-seminar-youssef-marzouk-mas sachusetts-institute-technology-shaffer-100/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-8216@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Title: Spectral Clustering for Dynamic Stochastic Block Mode l

\n\n

Abstract: One of the most common and crucial aspects o f many network data sets is the dependence of network link structure on ti me. In this work\, we extend the existing (static) nonparametric latent va riable model in the context of time-varying networks\, and thereby propos e a class of dynamic network models. For some special cases of these model s (namely the dynamic stochastic block model and dynamic degree corrected block model)\, which assume that there is a common clustering structure fo r all networks\, we consider the problem of identifying the common cluster ing structure. We propose two extensions of the (standard) spectral cluste ring method for the dynamic network models\, and give theoretical guarante e that the spectral clustering methods produce consistent community detect ion in case of both dynamic stochastic block model and dynamic degree-corr ected block model. The methods are shown to work under sufficiently mild conditions on the number of time snapshots to detect both associative and dissociative community structure\, even if all the individual networks ar e very sparse and most of the individual networks are below community dete ctability threshold. We reinforce the validity of the theoretical results via simulations too.

\n(Joint work with Shirshendu Chatterjee\, CUNY )

DTSTART;TZID=America/New_York:20161201T133000 DTEND;TZID=America/New_York:20161201T143000 SEQUENCE:0 SUMMARY:Seminar: Sharmodeep Bhattacharyya (Oregon State University) @ White head 304 URL:https://engineering.jhu.edu/ams/events/seminar-sharmodeep-bhattacharyya -oregon-state-university-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-7316@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT:http://www.math.jhu.edu/~data/ DESCRIPTION: DTSTART;TZID=America/New_York:20161207T150000 DTEND;TZID=America/New_York:20161207T160000 SEQUENCE:0 SUMMARY:Data Seminar: Ben Adcock (Simons Fraser University) @ Krieger 309 URL:https://engineering.jhu.edu/ams/events/data-seminar-ben-adcock-simons-f raser-university-shaffer-100/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-8240@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Title: Online and Random-order Load Balancing Simultaneously

\nAbstract: We consider the problem of online load balancing under lp-norms: sequential jobs need to be assigned to one of the machines and t he goal is to minimize the lp-norm of the machine loads. This generalizes the classical problem of scheduling for makespan minimization (case l_inft y) and has been thoroughly studied. We provide algorithms with simultaneou sly optimal* guarantees for the worst-case model as well as for the random -order (i.e. secretary) model\, where an arbitrary set of jobs comes in ra ndom order.

\nOne of the main components is a new algorithm with imp roved regret for Online Linear Optimization (OLO) over the non-negative ve ctors in the lq ball. Interestingly\, this OLO algorithm is also used to p rove a purely probabilistic inequality that controls the correlations aris ing in the random-order model\, a common source of difficulty for the anal ysis. A property that drives both our load balancing algorithms and our OL O algorithm is a smoothing of the the lp-norm that may be of independent i nterest.

DTSTART;TZID=America/New_York:20161208T133000 DTEND;TZID=America/New_York:20161208T143000 SEQUENCE:0 SUMMARY:Seminar: Marco Molinaro (Pontifical Catholic University of Rio de J aneiro) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-marco-molinaro-universit y-california-davis-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-8220@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:**Spatial-temporal modeling of the association between
air pollution exposures and birth outcomes: identifying critical exposure
windows**

Exposure to high levels of air pollution during pregnancy has been linked to increased probability of adverse birth outcom es. We consider statistical models for evaluating associations between pol lutants and birth outcomes\, taking into account multipollutant exposures\ , susceptible windows in pregnancy\, and variability in exposure over spac e and time. We consider geocoded vital records data from Texas as well as data from the National Birth Defects Prevention Study.

DTSTART;TZID=America/New_York:20170202T133000 DTEND;TZID=America/New_York:20170202T143000 SEQUENCE:0 SUMMARY:Wierman Lecture Series: Amy Herring (University of North Carolina C hapel Hill) @ Hodson 110 URL:https://engineering.jhu.edu/ams/events/wierman-lecture-series-amy-herri ng-university-north-carolina-chapel-hill-tba/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-9274@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT:http://www.math.jhu.edu/~data/ DESCRIPTION:\nDTSTART;TZID=America/New_York:20170208T150000 DTEND;TZID=America/New_York:20170208T160000 SEQUENCE:0 SUMMARY:Data Seminar: Kasso Okoudjou (University of Maryland\, College Park ) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/data-seminar-kasso-okoudjou-univ ersity-maryland-college-park-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-7404@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:

Title: Geometry\, Shapes and PDEs

\n\n

Abstract :

\nThe interest in right invariant metrics on the diffeomorphism gr oup is fueled by its relations to hydrodynamics. Arnold noted in 1966 that Euler’s equations\, which govern the motion of ideal\, incompressible flu ids\, can be interpreted as geodesic equations on the group of volume pres erving diffeomorphisms with respect to a suitable Riemannian metric. Since then other PDEs arising in physics have been interpreted as geodesic equa tions on manifold of mappings. Examples include Burgers’ equation\, the Kd V and Camassa-Holm equations or the Hunter-Saxton equation.

\nAnothe r important motivation for the study of Riemannian metrics on manifold of mappings can be found in its appearance in the field of shape analysis an d in particular in the eminent role of the diffeomorphism group in computa tional anatomy: the space of medical images is acted upon by the diffeomor phism group and differences between images are encoded by diffeomorphisms in the spirit of Grenander’s pattern theory. The study of anatomical shape s can be thus reduced to the study of the diffeomorphism group.

\nUs ing these observations as a starting point\, I will consider Riemannian me trics on spaces of mappings. I will discuss the local and global well-pose dness of the corresponding geodesic equation\, study the induced geodesic distance and present selected numerical examples of minimizing geodesics.< /p> DTSTART;TZID=America/New_York:20170209T133000 DTEND;TZID=America/New_York:20170209T143000 SEQUENCE:0 SUMMARY:Seminar: Martin Bauer (Florida State University) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-martin-bauer-florida-sta te-university-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-9278@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION: DTSTART;TZID=America/New_York:20170215T150000 DTEND;TZID=America/New_York:20170215T160000 SEQUENCE:0 SUMMARY:Data Seminar: Jerome Darbon (Brown University) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/data-seminar-jerome-darbon-brown -university-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-8440@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:

**TITLE:**

A local limit theorem for Qu ickSort key comparisons via multi-round smoothing

\n**ABSTRACT
:**

It is a well-known result\, due independently to Régnie r (1989) and Rösler (1991)\, that the number of key comparisons required b y the randomized sorting algorithm QuickSort to sort a list of n distinc t items (keys) satisfies a global distributional limit theorem. We resolv e an open problem of Fill and Janson from 2002 by using a multi-round smoo thing technique to establish the corresponding local limit theorem. (in pl ain text\; note that only the “n” in “sort a list of n distinct items” w ould be set in math mode in LaTeX)

\nThis is joint work with Béla Bo llobás and Oliver Riordan.

DTSTART;TZID=America/New_York:20170216T133000 DTEND;TZID=America/New_York:20170216T143000 SEQUENCE:0 SUMMARY:Seminar: Jim Fill (JHU) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-jim-fill-jhu-whitehead-3 04/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-8420@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION: DTSTART;TZID=America/New_York:20170223T133000 DTEND;TZID=America/New_York:20170223T143000 SEQUENCE:0 SUMMARY:Seminar: Laurent Younes (JHU) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-laurent-younes-jhu-white head-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-9442@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Link to the slides from Tom Loredo’s seminar- JHU17-HierBayesCosmicPopns

DTSTART;TZID=America/New_York:20170302T133000 DTEND;TZID=America/New_York:20170302T143000 SEQUENCE:0 SUMMARY:Seminar: Tom Loredo (Cornell University) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-tom-loredo-cornell-unive rsity-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-9470@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION: DTSTART;TZID=America/New_York:20170307T130000 DTEND;TZID=America/New_York:20170307T140000 LOCATION:Hodson 3rd Floor Lobby SEQUENCE:0 SUMMARY:HUSAM Presents Coffee Chat URL:https://engineering.jhu.edu/ams/events/husam-presents-coffee-chat/ X-COST-TYPE:free X-WP-IMAGES-URL:thumbnail\;https://engineering.jhu.edu/ams/wp-content/uploa ds/sites/44/2017/03/Coffee_Chat-002-300x139.jpg\;300\;139\,medium\;https:/ /engineering.jhu.edu/ams/wp-content/uploads/sites/44/2017/03/Coffee_Chat-0 02-300x139.jpg\;300\;139\,large\;https://engineering.jhu.edu/ams/wp-conten t/uploads/sites/44/2017/03/Coffee_Chat-002-300x139.jpg\;300\;139\,full\;ht tps://engineering.jhu.edu/ams/wp-content/uploads/sites/44/2017/03/Coffee_C hat-002-300x139.jpg\;300\;139 END:VEVENT BEGIN:VEVENT UID:ai1ec-9286@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT:http://www.math.jhu.edu/~data/ DESCRIPTION: DTSTART;TZID=America/New_York:20170308T150000 DTEND;TZID=America/New_York:20170308T160000 SEQUENCE:0 SUMMARY:Data Seminar: Matthew Hirn (Michigan State University) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/data-seminar-matthew-hirn-michig an-state-university-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-9262@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Title: Mean Field Games: theory and applications

\n< /p>\n

Abstract: We review the Mean Field Game paradigm introduced indepe ndently by Caines-Huang-Malhame and Lasry-Lyons ten years ago\, and we ill ustrate their relevance to applications with a few practical of examples ( bird flocking\, room exit\, systemic risk\, cyber-security\, …. ). We then review the probabilistic approach based on Forward-Backward Stochastic Di fferential Equations\, and we derive the Master Equation from a version o f the chain rule (Ito’s formula) for functions over flows of probability m easures. Finally\, motivated by the literature on economic models of bank runs\, we introduce mean field games of timing and discuss new results\, a nd some of the many remaining challenges.

DTSTART;TZID=America/New_York:20170309T133000 DTEND;TZID=America/New_York:20170309T143000 SEQUENCE:0 SUMMARY:Duncan Lecture Series: Rene Carmona (Princeton) @ Krieger 205 URL:https://engineering.jhu.edu/ams/events/duncan-lecture-series-rene-carmo na-princeton-tba/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-9266@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Title: Mean Field Games with Major and Minor Players: Theory and Numerics.

\n\n

Abstract: We present a (possibly) new for mulation of the mean field game problem in the presence of major and minor players\, and give new existence results for linear quadratic models and models with finite state spaces. We shall also provide numerical results i llustrating the theory and raising new challenging open problems.

DTSTART;TZID=America/New_York:20170310T133000 DTEND;TZID=America/New_York:20170310T143000 SEQUENCE:0 SUMMARY:Duncan Lecture Series: Rene Carmona (Princeton) @ Krieger 205 URL:https://engineering.jhu.edu/ams/events/duncan-lecture-series-rene-carmo na-princeton-tba-2/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-9483@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:HUSAM is hosting a professional event with Deloitte. The ev ent will be an overview of consulting at Deloitte. A panel of Deloitte pr actitioners will present on Deloitte’s BTA consulting track\, health analy tics\, and answer questions.

DTSTART;TZID=America/New_York:20170313T183000 LOCATION:Arellano Theater in Levering Hall SEQUENCE:0 SUMMARY:HUSAM Deloitte Event URL:https://engineering.jhu.edu/ams/events/husam-deloitte-event/ X-COST-TYPE:free X-INSTANT-EVENT:1 END:VEVENT BEGIN:VEVENT UID:ai1ec-9258@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Title: Nuke the Clouds: Using nuclear norm optimization to r emove clouds from satellite images

\nAbstract: We discuss how to use the nuclear norm and matrix factorization techniques to remove clouds fro m satellite images. The talk will focus on discussing the key properties and variational inequalities that is commonly used in minimizing convex fu nctions with a nuclear norm term. We will also contrast the convex formul ations with the corresponding rank constrained problems that are highly no n-convex\, but which are sometimes simpler to solve regardless. Finally w e will show a lot of examples/demos of how this is working in practice.

DTSTART;TZID=America/New_York:20170316T133000 DTEND;TZID=America/New_York:20170316T143000 SEQUENCE:0 SUMMARY:Seminar: Peder Olsen (IBM) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-peder-olsen-ibm-whitehea d-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-9290@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT:http://www.math.jhu.edu/~data/ DESCRIPTION: DTSTART;TZID=America/New_York:20170329T150000 DTEND;TZID=America/New_York:20170329T160000 SEQUENCE:0 SUMMARY:Data Seminar: Jason Eisner (JHU) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/data-seminar-jason-eisner-jhu-wh itehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-8224@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Heuristics for Network Revenue Management

\nWe conside r a network revenue management problem with customer choice and exogenous prices. Such problems are central in several applications including airlin e ticket pricing. Given the infeasibility of explicitly finding optimal po licies\, we study the performance of a class of heuristic policies. These heuristics periodically re-solve the deterministic linear program (DLP) th at results when all future random variables are replaced by their average values and implement the solutions in a probabilistic manner. We provide a n upper bound for the expected revenue loss under such policies when compa red to the optimal policy. Using this bound\, we construct a schedule of r e-solving times such that the resulting expected revenue loss is bounded b y a constant that is independent of the size of the problem.

\nJoint work with Stefanus Jasin at University of Michigan.

\nDTSTART;TZID=America/New_York:20170330T133000 DTEND;TZID=America/New_York:20170330T143000 SEQUENCE:0 SUMMARY:Seminar: Sunil Kumar (JHU) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-sunil-kumar-jhu-whitehea d-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-7408@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:

Energy Prices & Dynamic Games with Stochastic Demand

\nThe dramatic decline in oil prices\, from around $110 per barrel in June 2014 to around $30 in January 2016 highlights the importance of competitio n between different energy producers. Indeed\, the price drop has been pr imarily attributed to OPEC’s strategic decision (until very recently) not to curb its oil production in the face of increased supply of shale gas an d oil in the US\, which was spurred by the development of fracking technol ogy. Most dynamic Cournot models focus on supply-side factors\, such as in creased shale oil\, and random discoveries. However declining and uncertai n demand from China is a major factor driving oil price volatility. We stu dy Cournot games in a stochastic demand environment\, and present asymptot ic and numerical results\, as well as a modified Hotelling’s rule for game s with stochastic demand.

DTSTART;TZID=America/New_York:20170403T103000 DTEND;TZID=America/New_York:20170403T113000 SEQUENCE:0 SUMMARY:Seminar: Ronnie Sircar (Princeton University) @ Ames 234 URL:https://engineering.jhu.edu/ams/events/seminar-ronnie-sircar-princeton- university-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-9310@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION: DTSTART;TZID=America/New_York:20170405T150000 DTEND;TZID=America/New_York:20170405T160000 SEQUENCE:0 SUMMARY:Data Seminar: Afonso Bandeira (NYU) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/data-seminar-afonso-bandeira-nyu -whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-9478@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION: DTSTART;TZID=America/New_York:20170405T150000 DTEND;TZID=America/New_York:20170405T160000 SEQUENCE:0 SUMMARY:HUSAM Presents Coffee Chat URL:https://engineering.jhu.edu/ams/events/husam-presents-coffee-chat-2/ X-COST-TYPE:free X-WP-IMAGES-URL:thumbnail\;https://engineering.jhu.edu/ams/wp-content/uploa ds/sites/44/2017/03/Coffee_Chat-002-300x139.jpg\;300\;139\,medium\;https:/ /engineering.jhu.edu/ams/wp-content/uploads/sites/44/2017/03/Coffee_Chat-0 02-300x139.jpg\;300\;139\,large\;https://engineering.jhu.edu/ams/wp-conten t/uploads/sites/44/2017/03/Coffee_Chat-002-300x139.jpg\;300\;139\,full\;ht tps://engineering.jhu.edu/ams/wp-content/uploads/sites/44/2017/03/Coffee_C hat-002-300x139.jpg\;300\;139 END:VEVENT BEGIN:VEVENT UID:ai1ec-9302@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION: DTSTART;TZID=America/New_York:20170412T150000 DTEND;TZID=America/New_York:20170412T160000 SEQUENCE:0 SUMMARY:Data Seminar: Andrew Christlieb (Michigan State University) @ White head 304 URL:https://engineering.jhu.edu/ams/events/data-seminar-andrew-christlieb-m ichigan-state-university-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-9366@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Reciprocal Graphical Models for Integrati ve Gene Regulatory Network Analysis

\nConstructing gene regulatory networks is a fundamental task in systems biology. We introduce a Gaussian reciprocal graphical model for inference about gene regulatory relationships by integrating mRNA gene expression an d DNA level information including copy number and methylation. Data integr ation allows for inference on the directionality of certain regulatory rel ationships\, which would be otherwise indistinguishable due to Markov equi valence. Efficient inference is developed based on simultaneous equation < /span>models. Bayesian model selection technique s are adopted to estimate the graph structure. We illustrate our approach by simulations and two applications in ZODIAC pairwise gene interaction an alysis and colon adenocarcinoma pathway analysis.

DTSTART;TZID=America/New_York:20170413T133000 DTEND;TZID=America/New_York:20170413T143000 SEQUENCE:0 SUMMARY:Seminar: Peter Mueller (University of Texas) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-peter-mueller-university -texas-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-8444@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:**From solving PDEs to machine learning PDEs: **

**An odyssey in computational mathematics**

*George Em Karniadakis*

*Divis
ion of Applied Mathematics\, Brown University*

** Abstract:** In the last 30 years I have pursued the numerical sol
ution of partial differential equations (PDEs) using spectral and spectral
elements methods for diverse applications\, starting from deterministic P
DEs in complex geometries\, to stochastic PDEs for uncertainty quantificat
ion\, and to fractional PDEs that describe non-local behavior in disordere
d media and viscoelastic materials. More recently\, I have been working on
solving PDEs in a fundamentally different way. I will present a new parad
igm in solving linear and nonlinear PDEs from noisy measurements without t
he use of the classical numerical discretization. Instead\, we infer the s
olution of PDEs from noisy data\, which can represent measurements of vari
able fidelity. The key idea is to encode the structure of the PDE into pri
or distributions and train Bayesian nonparametric regression models on ava
ilable noisy data. The resulting posterior distributions can be used to pr
edict the PDE solution with quantified uncertainty\, efficiently identify
extrema via Bayesian optimization\, and acquire new data via active learni
ng. Moreover\, I will present how we can use this new framework to learn P
DEs from noisy measurements of the solution and the forcing terms.

\n

**Bio**: George Karniadakis received his S.M. and
Ph.D. from Massachusetts Institute of Technology. He was appointed Lecture
r in the Department of Mechanical Engineering at MIT in 1987 and subsequen
tly he joined the Center for Turbulence Research at Stanford / Nasa Ames.
He joined Princeton University as Assistant Professor in the Department of
Mechanical and Aerospace Engineering and as Associate Faculty in the Prog
ram of Applied and Computational Mathematics. He was a Visiting Professor
at Caltech in 1993 in the Aeronautics Department and joined Brown Universi
ty as Associate Professor of Applied Mathematics in the Center for Fluid M
echanics in 1994. After becoming a full professor in 1996\, he continues t
o be a Visiting Professor and Senior Lecturer of Ocean/Mechanical Engineer
ing at MIT. He is a Fellow of the Society for Industrial and Applied Mathe
matics (SIAM\, 2010-)\, Fellow of the American Physical Society (APS\, 200
4-)\, Fellow of the American Society of Mechanical Engineers (ASME\, 2003-
) and Associate Fellow of the American Institute of Aeronautics and Astron
autics (AIAA\, 2006-). He received the Ralf E Kleinman award from SIAM (20
15)\, the J. Tinsley Oden Medal (2013)\, and the CFD award (2007) by the U
S Association in Computational Mechanics. His h-index is 79 and he has bee
n cited over 32\,500 times.

DTSTART;TZID=America/New_York:20170420T133000 DTEND;TZID=America/New_York:20170420T143000 SEQUENCE:0 SUMMARY:Seminar: George Karniadakis (Brown University) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-george-karniadakis-brown -university-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-9298@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:

Solving Fredholm integrals from incomplete
measurements

\nWe present an algorithm to solve Fredholm integrals of the first kind with tensor product structures\, from a limited number of measurements with the goal of using this method to accelerate Nuclear M agnetic Resonance (NMR) acquisition. This is done by incorporating compres sive sampling type arguments to fill in the missing measurements using a p riori knowledge of the structure of the data. In the first step\, we recov er a compressed data matrix from measurements that form a tight frame\, an d establish that these measurements satisfy the restricted isometry proper ty (RIP). In the second step\, we solve the zeroth-order regularization mi nimization problem using the Venkataramanan-Song-Huerlimann algorithm. We demonstrate the performance of this algorithm on simulated and real data a nd we compare it with other sampling techniques. Our theory applied to bot h 2D and multidimensional NMR.

DTSTART;TZID=America/New_York:20170426T150000 DTEND;TZID=America/New_York:20170426T160000 SEQUENCE:0 SUMMARY:Data Seminar: Wojciech Czaja (University of Maryland\, College Park ) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/data-seminar-wojciech-czaja-univ ersity-maryland-college-park-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-8416@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Title: Parametrization of discrete optimization problems\, s ubdeterminants and matrix-decomposition

\n\n

Abstract:

\n< p>The central goal of this talk is to identify parameters that explain the complexity of Integer linear programming defined as follows:\nLet P be a polyhedron. Determine an integral point in P that maximizes a linea r function.

\n\n

It is obvious that the number of integer var iables is such a parameter.

\nHowever\, in view of applications in v ery high dimensions\, the question emerges whether we need to treat all va riables as integers? In other words\, can we reparametrize integer program s with significantly less integer variables?

\n\n

A second mu ch less obvious parameter associated with an integer linear program is the number Delta defined as the maximum absolute value of any square submatri x of a given integral matrix A with m rows and n columns.

\nThis lea ds us to the important open question whether we can solve integer linear p rogramming in a polynomial running time in Delta and the instance size?

\n\n

Regarding the first question\, we exhibit a variety of exa mples that demonstrate how integer programs can be reformulated using much less integer variables. To this end\, we introduce a generalization of to tal unimodularity called the affine TU-dimension of a matrix and study rel ated theory and algorithms for determining the affine TU-dimension of a ma trix.

\n\n

Regarding the second question\,

\nwe present several new results that illustrate why Delta is an important parameter a bout the complexity of integer linear programs associated with a given mat rix A.

\nIn particular\, in the nondegenerate case integer linear pr ograms with any constant value Delta can be solved in polynomial time.

\nThis extends earlier results of Veselov and Chirkov.

DTSTART;TZID=America/New_York:20170427T133000 DTEND;TZID=America/New_York:20170427T143000 SEQUENCE:0 SUMMARY:Seminar: Robert Weismantel (ETH Zurich) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-robert-weismantel-eth-zu rich-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-9270@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Title: Vo l\, Skew\, and Smile Trading

\nAbstract: We use dynamically traded portfolios of opt ions to bet on either the quadratic variation of log price\, or on the rea lized co-variation of log price with log implied vol\, or on the quadratic variation of implied vol. Our bets lead to precise financial meanings for the level\, slo pe\, and curvature of implied variance in moneyness.

\nDTSTART;TZID=America/New_York:20170504T133000 DTEND;TZID=America/New_York:20170504T143000 SEQUENCE:0 SUMMARY:Seminar: Peter Carr (NYU) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-peter-carr-nyu-whitehead -304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-9294@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT:http://www.math.jhu.edu/~data/ DESCRIPTION: DTSTART;TZID=America/New_York:20170510T150000 DTEND;TZID=America/New_York:20170510T160000 SEQUENCE:0 SUMMARY:Data Seminar: Rachel Ward (University of Texas\, Austin) @ Whitehea d 304 URL:https://engineering.jhu.edu/ams/events/data-seminar-rachel-ward-univers ity-texas-austin-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10353@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:

\n

Title: Introduction to the Financial Mathemat ics Seminar

\n\n

Abstract:

\nThe seminar will have two parts:

\nPart I) Daniel Naiman will introduce the Financial Mathemat ics seminar.

\nPart II) Sonjala Williams will speak about networking and job search strategies.

DTSTART;TZID=America/New_York:20170905T133000 DTEND;TZID=America/New_York:20170905T150000 LOCATION:Shaffer 101 SEQUENCE:0 SUMMARY:Financial Mathematics Seminar: Daniel Naiman and Sonjala Williams ( JHU) @ Shaffer 101 URL:https://engineering.jhu.edu/ams/events/introduction-to-the-financial-ma thematics-seminar/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10398@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Dr. Dave Schrader will be speaking at Sports Analytics Club’ s first meeting of the year on Monday\, September 11 at 7pm in Schafer 3. Dr. Schrader is an expert in the field of sports analytics\, and has a wid e breadth of experience in industry and in speaking to college students ab out the field. Individuals with all levels of experience in sports analyti cs are welcome to attend\, and the talk should give a good overview of how sports analytics are currently being used and how students can get involv ed themselves.

\n\n

The Dr. Schrader’s talk is entitled “The Golden Age of Sports Analytics\,” and it will cover the following topics:< /p>\n

- \n
- What’s happening around the world to collect and analyze da ta for recruiting\, player development\, game planning\, and injury preven tion? \n
- How are analytics being used to evaluate and improve busi ness operations – ticket pricing\, sales\, sponsorships? \n
- What an alytics are the leading pro teams and leagues using for basketball\, baseb all\, football\, hockey\, and soccer? \n
- How quickly are analytics being adopted at the college level? Who is leading? What are they doing?\n
- How can other parts of the university\, like the business school or computer science departments\, collaborate with sports programs to prov ide analytics for teams? What are good “Moneyball” projects to launch? W hat have other schools done? \n
- Where can you get more information? What to read? What conferences to attend? \n

Title: No equations\, no variables\, no parameters\, no spac e\, no time: Data and the computational modeling of complex/multiscale sys tems

\nAbstract: Obtaining predictive dynamical equations from data lies at the heart of science and engineering modeling\, and is the linchpi n of our technology. In mathematical modeling one typically progresses fro m observations of the world (and some serious thinking!) first to equation s for a model\, and then to the analysis of the model to make predictions. Good mathematical models give good predictions (and inaccurate ones do no t) – but the computational tools for analyzing them are the same: algorith ms that are typically based on closed form equations. While the skeleton o f the process remains the same\, today we witness the development of mathe matical techniques that operate directly on observations -data-\, and appe ar to circumvent the serious thinking that goes into selecting variables a nd parameters and deriving accurate equations. The process then may appear to the user a little like making predictions by “looking in a crystal bal l”. Yet the “serious thinking” is still there and uses the same -and some new- mathematics: it goes into building algorithms that “jump directly” fr om data to the analysis of the model (which is now not available in closed form) so as to make predictions. Our work here presents a couple of effor ts that illustrate this “new” path from data to predictions. It really is the same old path\, but it is travelled by new means.

\nRelated pap
ers:

\nParsimonious Representation of Nonlinear Dynamical Systems thr
ough Manifold Learning: a Chemotaxis Case Study

\nAn Equal Space for
Complex Data with Unknown Internal Order: Observability\, Gauge Invariance
and Manifold Learning Kevrekidis

Symmetry\, Temporal Information\, and Succinct Representation of Random Graph Structures

\nI will dis cuss mathematical aspects of my recent work on two related problems at the intersection of random graphs and information theory: (i) node order infe rence – for a dynamic random graph model\, determine the extent to which t he order in which nodes arrived can be inferred from the graph structure\, and (ii) source coding of structures – for a given graph model\, exhibit an efficiently computable and invertible mapping from unlabeled graphs to bit strings with minimum possible expected code length. Both problems are connected to the study of the symmetries of the graph model\, as well as a nother combinatorial quantity – the typical number of feasible labeled rep resentatives of a given structure. I will focus on the case of the prefere ntial attachment model\, for which we are able to give a (nearly) complete characterization of the behavior of the size of the automorphism group\, as well as a provably asymptotically optimal algorithm for (ii)\, and opti mal estimators for certain natural formulations of (i).

\nDTSTART;TZID=America/New_York:20170914T133000 DTEND;TZID=America/New_York:20170914T143000 SEQUENCE:0 SUMMARY:AMS Seminar: Abram Magner (University of Illinois) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-abram-mager-illinois-uni versity-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10414@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:

Title: Semiparametric spectral modeling of the Drosophila c onnectome

\nAbstract: We present semiparametric spectral modeling o f the complete larval Drosophila mushroom body connectome. Motivated by a thorough exploratory data analysis of the network via Gaussian mixture mod eling (GMM) in the adjacency spectral embedding (ASE) representation space \, we introduce the latent structure model (LSM) for network modeling and inference. LSM is a generalization of the stochastic block model (SBM) and a special case of the random dot product graph (RDPG) latent position mod el\, and is amenable to semiparametric GMM in the ASE representation space . The resulting connectome code derived via semiparametric GMM composed wi th ASE captures latent connectome structure and elucidates biologically re levant neuronal properties.

\nRelated papers:

\nThe complete c
onnectome of a learning and memory center in an insect brain

\nA cons
istent adjacency spectral embedding for stochastic blockmodel graphs

\nA limit theorem for scaled eigenvectors of random dot product graphs

\nLimit theorems for eigenvectors of the normalized Laplacian for random
graphs

TITLE – On optimizing a submodular utility function

\nABSTRACT – This talk has two related parts. Part one is on the maximizatio n of a particular submodular utility function\, whereas part two is on its minimization. Both problems arise naturally in combinatorial optimization with risk aversion\, including estimation of project duration with stocha stic task times\, in reliability models\, multinomial logit models\, compe titive facility location\, combinatorial auctions\, as well as in portfoli o optimization.

\nPart 1: Given a monotone concave univariate functi on g\, and two vectors c and d\, we consider the discrete optimization pro blem of finding a vertex of a polytope maximizing the utility function c’x + g(d’x). The problem is NP-hard for any strictly concave function g even for simple polytopes\, such as the uniform matroid\, assignment and path polytopes. We give a 1/2-approximation algorithm for it and improvements f or special cases\, where g is the square root\, log utility\, negative exp onential utility and multinomial logit probability function. In particular \, for the square root function\, the approximation ratio improves to 4/5. Although the worst case bounds are tight\, computational experiments indi cate that the approximation algorithm finds solutions within 1-2% optimali ty gap for most of the instances very quickly and can be considerably fast er than the existing alternatives.

\nPart 2: We consider a mixed 0-1 conic quadratic optimization problem with indicator variables arising in mean-risk optimization. The indicator variables are often used to model no n-convexities such as fixed charges or cardinality constraints. Observing that the problem reduces to a submodular function minimization for its bin ary restriction\, we derive three classes of strong convex valid inequalit ies by lifting the polymatroid inequalities on the binary variables. Compu tational experiments demonstrate the effectiveness of the inequalities in strengthening the convex relaxations and\, thereby\, improving the solutio n times for mean-risk problems with fixed charges and cardinality constrai nts significantly.

DTSTART;TZID=America/New_York:20170921T133000 DTEND;TZID=America/New_York:20170921T143000 SEQUENCE:0 SUMMARY:AMS Seminar: Alper Atamturk (Berkeley University) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/seminar-alper-atamturk-berkeley- university-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10417@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Title: Frames — two case studies: ambiguity and uncertainty< /p>\n

Abstract: The theory of frames is an essential concept for dealing with signal representation in noisy environments. We shall examine the th eory in the settings of the narrow band ambiguity function and of quantum information theory. For the ambiguity function\, best possible estimates a re derived for applicable constant amplitude zero autocorrelation (CAZAC) sequences using Weil’s solution of the Riemann hypothesis for finite field s. In extending the theory to the vector-valued case modelling multi-senso r environments\, the definition of the ambiguity function is characterized by means of group frames. For the uncertainty principle\, Andrew Gleason’ s measure theoretic theorem\, establishing the transition from the lattice interpretation of quantum mechanics to Born’s probabilistic interpretatio n\, is generalized in terms of frames to deal with uncertainty principle i nequalities beyond Heisenberg’s. My collaborators are Travis Andrews\, Rob ert Benedetto\, Jeffrey Donatelli\, Paul Koprowski\, and Joseph Woodworth.

\nRelated papers:

\nSuper-resolution by means of Beurling min
imal extrapolation

\nGeneralized Fourier frames in terms of balayage<
br />\nUncertainty principles and weighted norm inequalities

\nA fram
e reconstruction algorithm with applications to magnetric resonance imagin
g

\nFrame multiplication theory and a vector-valued DFT and ambiguity
functions

~~Title: An improved approach to calibrating misspecified mathematical models~~

Abstract: We consider the problem of calibrating misspecified mathematical model s using experimental data. To compensate for the misspecification of the m odel\, a discrepancy function is usually included and modeled via a Gaussi an stochastic process (GaSP)\, leading to better results of prediction. Th e calibration parameters in the model\, however\, sometimes become unident ifiable and the calibrated model fits the experimental data poorly as a co nsequence. In this work\, we propose the scaled Gaussian stochastic proces s (S-GaSP)\, a novel stochastic process for calibration and prediction. Th is new approach bridges the gap between two predominant methods\, namely t he $L_2$ calibration and GaSP calibration. A computationally feasible appr oach is introduced for this new model under the Bayesian paradigm. The S-G aSP model not only provides a general framework for calibration\, but also enables the calibrated mathematical model to predict well regardless of t he discrepancy function. Simulation examples are provided and real example s using satellite images to calibrate the model for volcanic hazard are st udied to illustrate the connections and differences between this new model and other previous approaches.

\nDTSTART;TZID=America/New_York:20170928T133000 DTEND;TZID=America/New_York:20170928T143000 SEQUENCE:0 SUMMARY:AMS Seminar: Mengyang Gu (JHU) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/ams-seminar-mengyang-gu-jhu-whit ehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10421@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:

Title: Data-driven discovery of governing equations and phys ical laws

\nAbstract: The emergence of data methods for the sciences in the last decade has been enabled by the plummeting costs of sensors\, computational power\, and data storage. Such vast quantities of data affor d us new opportunities for data-driven discovery\, which has been referred to as the 4th paradigm of scientific discovery. We demonstrate that we ca n use emerging\, large-scale time-series data from modern sensors to direc tly construct\, in an adaptive manner\, governing equations\, even nonline ar dynamics\, that best model the system measured using modern regression techniques. Recent innovations also allow for handling multi-scale physics phenomenon and control protocols in an adaptive and robust way. The overa ll architecture is equation-free in that the dynamics and control protocol s are discovered directly from data acquired from sensors. The theory deve loped is demonstrated on a number of canonical example problems from physi cs\, biology and engineering.

\nRelated papers:

\nDiscovering
governing equations from data by sparse identification of nonlinear dynami
cal systems

\nData-driven discovery of partial differential equations

\nChaos as an intermittently forced linear system

Title: Multidimensional wavelet signal denoising via adaptiv e random partitioning

\n\n

Abstract: Traditional statistical wavelet analysis usually focuses on modeling the wavelet coefficients unde r a given\, predetermined wavelet transform. Such analysis may quickly los e efficiency in multivariate problems under traditional multivariate wavel et transforms\, which are symmetric with respect to the dimensions\, as pr edetermined wavelet transforms cannot adaptively exploit the energy distri bution in a problem-specific manner. We introduce a principled probabilist ic framework for incorporating such adaptivity by (i) representing multiva riate functions using one-dimensional (1D) wavelet transforms applied to a permuted version of the original function\, and (ii) placing a hyperprior on the corresponding permutation. Such a representation can achieve subst antially better energy concentration in the wavelet coefficients and highl y scalable inference algorithms. In particular\, when combined with the Ha ar basis\, we obtain the exact Bayesian inference analytically through a r ecursive message passing algorithm with a computational complexity that sc ales linearly with sample size. In addition\, we propose a sequential Mont e Carlo (SMC) inference algorithm for other wavelet bases using the exact Haar solution as the proposal. We demonstrate that with this framework eve n simple 1D Haar wavelets can achieve excellent performance in both 2D and 3D image reconstruction via numerical experiments\, outperforming state-o f-the-art multidimensional wavelet-based methods especially in low signal- to-noise ratio settings\, at a fraction of the computational cost.

\n\n

This is a joint work with Li Ma.

DTSTART;TZID=America/New_York:20171005T133000 DTEND;TZID=America/New_York:20171005T143000 SEQUENCE:0 SUMMARY:AMS Seminar: Meng Li (Rice University) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/ams-seminar-meng-li-rice-univers ity-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10566@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:* Title*: Discrete Nonlinear Optimiza
tion by State-Space Decompositions

\n

**Abstract<
/strong>****: In this talk we will discuss a decomposition approach for b
inary optimization problems with nonlinear objectives and linear constrain
ts. Our methodology relies on the partition of the objective function into
separate low-dimensional dynamic programming (DP) models\, each of which
can be equivalently represented as a shortest-path problem in an underlyin
g state transition graph. We show that the associated transition graphs ca
n be related by a mixed-integer linear program (MILP) so as to produce exa
ct solutions to the original nonlinear problem. To address DPs with large
state spaces\, we present a general relaxation mechanism which dynamically
aggregates states during the construction of the transition graphs. The r
esulting MILP provides both lower and upper bounds to the nonlinear functi
on\, and may be embedded in branch-and-bound procedures to find provably o
ptimal solutions. We describe how to specialize our technique for structur
ed objectives (e.g.\, submodular functions) and consider three problems ar
ising in revenue management\, portfolio optimization\, and healthcare. Num
erical studies indicate that the proposed technique often outperforms stat
e-of-the-art approaches by orders of magnitude in these applications.**

**Title:** Discussion of quantitative careers i
n private banking

\n

**Abstract:** TBA

Title: Data assimilation with stochastic model reduction

\nAbstract: In weather and climate prediction\, data assimilation combi nes data with dynamical models to make prediction\, using ensemble of solu tions to represent the uncertainty. Due to limited computational resources \, reduced models are needed and coarse-grid models are often used\, and t he effects of the subgrid scales are left to be taken into account. A majo r challenge is to account for the memory effects due to coarse graining wh ile capturing the key statistical-dynamical properties. We propose to use nonlinear autoregression moving average (NARMA) type models to account for the memory effects\, and demonstrate by examples that the resulting NARMA type stochastic reduced models can capture the key statistical and dynami cal properties and therefore can improve the performance of ensemble predi ction in data assimilation. The examples include the Lorenz 96 system (whi ch is a simplified model of the atmosphere) and the Kuramoto-Sivashinsky e quation of spatiotemporally chaotic dynamics.

\nRelated papers:

\nDiscrete approach to stochastic parametrization and dimension reductio
n in nonlinear dynamics

\nAccounting for model error from unresolved
scales in ensemble Kalman filters by stochastic parametrization

**Title: **Financial Contagion and Systemic Ris
k

**Abstract: **Financial contagion occurs when the d
istress of one bank jeopardizes the health of other financial firms\, and
can ultimately spread to the real economy. The spread of defaults in the f
inancial system can occur due to both local connections\, e.g.\, contractu
al obligations\, and global connections\, e.g.\, through the prices of ass
ets due to mark-to-market valuation. As evidenced by the 2007-2009 financi
al crisis\, the cost of a systemic event is tremendous\, thus requiring a
detailed look at the contributing factors. In this talk\, we will detail t
he local contagion model of Eisenberg and Noe (2001). However\, in utilizi
ng this model\, central bankers and regulators often must estimate the int
erbank liabilities because complete information on bilateral obligations i
s rarely available. This estimation can introduce errors to the level of f
inancial contagion and risk in the system. We will consider a sensitivity
analysis of the Eisenberg-Noe model to determine the size of these potenti
al estimation errors.

Title: “Hyper-Molecules” in Cryo-Electron Microscopy (cryo-E M)

\nAbstract: Cryo-EM is an imaging technology that is revolutioniz ing structural biology\; the Nobel Prize in Chemistry 2017 was recently aw arded to Jacques Dubochet\, Joachim Frank and Richard Henderson “for devel oping cryo-electron microscopy for the high-resolution structure determina tion of biomolecules in solution”. Cryo-electron microscopes produce a lar ge number of very noisy two-dimensional projection images of individual fr ozen molecules. Unlike related tomography methods\, such as computed tomog raphy (CT)\, the viewing direction of each image is unknown. The unknown d irections\, together with extreme levels of noise and additional technical factors\, make the determination of the structure of molecules challengin g. Unlike other structure determination methods\, such as x-ray crystallog raphy and nuclear magnetic resonance (NMR)\, cryo-EM produces measurements of individual molecules and not ensembles of molecules. Therefore\, cryo- EM could potentially be used to study mixtures of different conformations of molecules. While current algorithms have been very successful at analyz ing homogeneous samples\, and can recover some distinct conformations mixe d in solutions\, the determination of multiple conformations\, and in part icular\, continuums of similar conformations (continuous heterogeneity)\, remains one of the open problems in cryo-EM. I will discuss the “hyper-mol ecules” approach to continuous heterogeneity\, and the numerical tools and analysis methods that we are developing in order to recover such hyper-mo lecules.

DTSTART;TZID=America/New_York:20171018T150000 DTEND;TZID=America/New_York:20171018T160000 SEQUENCE:0 SUMMARY:Data Science Seminar: Roy Lederman (Princeton University) @ Hodson 203 URL:https://engineering.jhu.edu/ams/events/data-science-seminar-roy-lederma n-princeton-university-hodson-203/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10484@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Title: Optimization with Polyhedral Constraints

\n\n

Abstract:

\nA two-phase strategy is presented for solving an o ptimization problem whose feasible set is a polyhedron. Phase one is the g radient projection algorithm\, while phase two is essentially any algorith m for solving a linearly constrained optimization problem. Using some simp le rules for branching between the two phases\, it is shown\, under suitab le assumptions\, that only the linearly constrained optimization algorithm is performed asymptotically. Hence\, the asymptotic convergence behavior of the two phase algorithm coincides with the convergence behavior of the linearly constrained optimizer. Numerical results are presented using CUTE test problems\, a recently developed algorithm for projecting a point ont o a polyhedron\, and the conjugate gradient algorithm for the linearly con strained optimizer.

DTSTART;TZID=America/New_York:20171019T133000 DTEND;TZID=America/New_York:20171019T143000 SEQUENCE:0 SUMMARY:AMS Seminar: Bill Hager (University of Florida) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/ams-seminar-bill-hager-universit y-florida-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10432@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:TBA

DTSTART;TZID=America/New_York:20171025T150000 DTEND;TZID=America/New_York:20171025T160000 SEQUENCE:0 SUMMARY:Data Science Seminar: John Harlim (Penn State University) @ Hodson 203 URL:https://engineering.jhu.edu/ams/events/data-science-seminar-john-harlim -penn-state-university-hodson-203/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10487@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Title: Coherence in Statistical Modeling of Networks

\n< p>Abstract:\nGeorge Box famously said\, “All models are wrong\, but some are useful.”

\nClassic texts define a statistical model as “a set of distributions on the sample space” (Cox and Hinkley\, 1976\; Lehman \, 1983\; Barndorff-Nielson and Cox\, Bernardo and Smith\, 1994).

\nMotivated by some longstanding questions in the analysis of network data\, I will examine both of these statements\, first from a general point of v iew\, and then in the context of some recent developments in network analy sis.

\nThe confusion caused by these statements is clarified by the realization that the definition of statistical model must be refined — it must be more than just a set. With this\, the ambiguity in Box’s statemen t — e.g.\, what determines whether a model is ‘wrong’ or ‘useful’? — can b e clarified by a logical property that I call ‘coherence’. After clarific ation\, a model is deemed useful as long as it is coherent\, i.e.\, infere nces from it ‘make sense’.

\nI will then discuss some implications f or the statistical modeling of network data.

DTSTART;TZID=America/New_York:20171026T133000 DTEND;TZID=America/New_York:20171026T143000 SEQUENCE:0 SUMMARY:AMS Seminar: Harry Crane ( Rutgers University) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/ams-seminar-harry-crane-rutgers- university-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10680@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:**Title:** Functional central limit theorems fo
r rough volatility models

**Abstract: **We extend Don
sker’s approximation of Brownian motion to fractional Brownian motion with
any Hurst exponent (including the ’rough’ case H < 1/2)\, and Volterra-li
ke processes. Some of the most relevant consequences of our ‘rough Donsker
(rDonsker) Theorem’ are convergence results (with rates) for discrete app
roximations of a large class of rough models. This justifies the validity
of simple and easy-to-implement Monte-Carlo methods\, for which we provide
detailed numerical recipes. We test these against the current benchmark h
ybrid scheme of Bennedsen\, Lunde\, and Pakkanen and find remarkable agree
ment (for a large range of values of H). This rDonsker Theorem further pro
vides a weak convergence proof for the hybrid scheme itself\, and allows t
o construct binomial trees for rough volatility models\, the first availab
le scheme (in the rough volatility context) for early exercise options suc
h as American or Bermudan. The talk is based on joint work with B. Horvath
and A. Muguruza.

** **

Title: Some matrix problems in quantum information science\n

Abstract:

\nIn this talk\, we present some matrix results and techniques in

\nsolving certain optimization problems arising in quantum informat
ion

\nscience.

No quantum mechanics background is requ ired.

\nDTSTART;TZID=America/New_York:20171101T133000 DTEND;TZID=America/New_York:20171101T143000 SEQUENCE:0 SUMMARY:AMS Seminar: Chi-Kwong Li (Williams and Mary University\, IQC) @ Bl oomberg 274 URL:https://engineering.jhu.edu/ams/events/ams-seminar-chi-kwong-li-william s-mary-university-bloomberg-274/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10686@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES;LANGUAGE=en-US:Financial Mathematics Seminar CONTACT: DESCRIPTION:

**Title: **Data scientist at Kensho working foc
using on natural language processing

Title: Emergent behavior in self-organized dynamics: from co nsensus to hydrodynamic flocking

\nAbstract: We discuss several firs t- and second-order models encountered in opinion and flocking dynamics. T he models are driven by different “rules of engagement”\, which quantify h ow each member interacts with its immediate neighbors. We highlight the ro le of geometric vs. topological neighborhoods and distinguish between loca l and global interactions\, while addressing the following two related que stions. (i) How local rules of interaction lead\, over time\, to the emerg ence of consensus\; and (ii) How the flocking behavior of large crowds cap tured by their hydrodynamic description.

DTSTART;TZID=America/New_York:20171108T150000 DTEND;TZID=America/New_York:20171108T160000 SEQUENCE:0 SUMMARY:Data Science Seminar: Eitan Tadmor (University of Maryland ) @ Hods on 203 URL:https://engineering.jhu.edu/ams/events/data-science-seminar-eitan-tadmo r-university-maryland-hodson-203/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10494@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Title: Market Efficiency with Micro and Macro Information\n

Abstract:

\nWe propose a tractable\, multi-security model in which investors allocate information processing capacity to acquire micro information about individual stocks and/or macro information about an inde x fund. Investors solve optimal portfolio selection and information alloca tion problems. In equilibrium\, all investors are of one of three types: m icro informed\, macro informed\, or uninformed. We investigate the implica tions for price efficiency and find an endogenous bias toward micro effici ency: over a range of parameter values prices will be more informative abo ut micro than macro fundamentals. We explore the model’s implications for the cyclicality of investor information choices\, for systematic and idios yncratic return volatility\, and for excess covariance and volatility. Thi s is joint work with Harry Mamaysky.

\n\n

\n
DTSTART;TZID=America/New_York:20171109T133000
DTEND;TZID=America/New_York:20171109T143000
SEQUENCE:0
SUMMARY:AMS Seminar: Paul Glasserman (Columbia University) @ Whitehead 304
URL:https://engineering.jhu.edu/ams/events/ams-seminar-paul-glasserman-colu
mbia-university-whitehead-304/
X-COST-TYPE:free
END:VEVENT
BEGIN:VEVENT
UID:ai1ec-10663@engineering.jhu.edu/ams
DTSTAMP:20180317T041026Z
CATEGORIES:
CONTACT:
DESCRIPTION:No quantum mechanics
background is required.

\nWhat geometries can we learn from data?

\nIn the field of manifold learning\, the foundational theoretical results of Coifman an d Lafon (Diffusion Maps\, 2006) showed that for data sampled near an embed ded manifold\, certain graph Laplacian constructions are consistent estima tors of the Laplace-Beltrami operator on the underlying manifold. Since th ese operators determine the Riemannian metric\, they completely describe t he geometry of the manifold (as inherited from the embedding). It was late r shown that different kernel functions could be used to recover any desir ed geometry\, at least in terms of pointwise estimation of the associated Laplace-Beltrami operator. In this talk I will first briefly review the ab ove results and then introduce new results on the spectral convergence of these graph Laplacians. These results reveal that not all geometries are a ccessible in the stronger spectral sense. However\, when the data set is s ampled from a smooth density\, there is a natural conformally invariant ge ometry which is accessible on all compact manifolds\, and even on a large class of non-compact manifolds. Moreover\, the kernel which estimates this geometry has a very natural construction which we call Continuous k-Neare st Neighbors (CkNN).

DTSTART;TZID=America/New_York:20171115T150000 DTEND;TZID=America/New_York:20171115T160000 SEQUENCE:0 SUMMARY:Data Science Seminar: Tyrus Berry (George Mason University) @ Hodso n 203 URL:https://engineering.jhu.edu/ams/events/data-science-seminar-tyrus-berry -george-mason-university-hodson-203/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10496@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:**Title:** Transaction Clock\, Stochastic Time
Changes and Stochastic Volatility

**Abstract**: The
first part of the talk will establish that\, by No Arbitrage\, the log – p
rice process of a stock has to be a time-changed Brownian motion under the
physical probability measure. Aggregate volume and number of trades are e
mpirically tested as possible drivers of the stochastic clock allowing one
to recover normality of stock returns.

The second part of the tal k will show how stochastic volatility can be represented through a stochas tic time change\, outside the stochastic differential equations classicall y used for volatility in a number of founding models in Finance. This repr esentation is particularly useful if one wishes to choose a Levy process ( outside Brownian motion) for the stock log- price\, as independent increme nts are contradicted by volatility clustering observed in financial market s. The CGMY process with stochastic volatility will be provided as an exam ple.

DTSTART;TZID=America/New_York:20171116T133000 DTEND;TZID=America/New_York:20171116T143000 SEQUENCE:0 SUMMARY:AMS Seminar: Helyette Geman (JHU) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/ams-seminar-eric-xing-carnegie-m ellon-university-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10603@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Join alumni\, faculty\, and students for a Financial Mathema tics alumni reunion\, including speakers\, networking event\, and happy ho ur. Food and drink will be served. (Informal Happy hour to follow)

\n< p> \nRegister @ https://jhu.us6. list-manage.com/track/click?u=40512314224886c4ca8b856c2&id=836eea02d4&e=19 07de3a2c

\n\n

If you have any questions\, please contact Sonjala Williams @ sonjala@jhu.edu

\n\n

DTSTART;TZID=America/New_York:20171118T120000 DTEND;TZID=America/New_York:20171118T160000 SEQUENCE:0 SUMMARY:JHU Whiting School of Engineering AMS- Financial Math Alumni Reunio n @ Glass Pavilion URL:https://engineering.jhu.edu/ams/events/jhu-whiting-school-engineering-a ms-financial-math-alumni-reunion-glass-pavilion/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10664@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:

TBA

DTSTART;TZID=America/New_York:20171129T150000 DTEND;TZID=America/New_York:20171129T160000 SEQUENCE:0 SUMMARY:Data Science Seminar: Yingzhou Li (Duke University) @ Hodson 203 URL:https://engineering.jhu.edu/ams/events/data-science-seminar-yingzhou-li -duke-university-hodson-203/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10552@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:**Title:** Distributed Synchronization in Engin
eering Networks

**Abstract:**

This talk prese nts a systematic study of synchronization on distributed (networked) syste ms that spans from theoretical modeling and stability analysis to distribu ted controller design\, implementation and verification. We first focus on developing a theoretical foundation for synchronization of networked osci llators. We study how the interaction type (coupling) and network configur ation (topology) affect the behavior of a population of heterogeneous coup led oscillators. Unlike existing literature that restricts to specific sce narios\, we show that phase consensus (common phase value) can be achieved for arbitrary network topologies under very general conditions on the osc illators’ model.

\nWe then focus on more practical aspects of synchr onization on computer networks. Unlike existing solutions that tend to rel y on expensive hardware to improve accuracy\, we provide a novel algorithm that reduces jitter by synchronizing networked computers without estimati ng the frequency difference between clocks (skew) or introducing offset co rrections. We show that a necessary and sufficient condition on the networ k topology for synchronization (in the presence of noise) is the existence of a unique leader in the communication graph. A Linux-based implementati on on a cluster of IBM BladeCenter servers experimentally verifies that th e proposed algorithm outperforms well-established solutions and that loops can help reduce jitter.

\nDTSTART;TZID=America/New_York:20171130T133000 DTEND;TZID=America/New_York:20171130T143000 SEQUENCE:0 SUMMARY:AMS Seminar: Enrique Mallada Garcia (JHU) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/ams-seminar-enrique-mallada-garc ia-jhu-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10832@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:

The Graduate Career Advisor for Financial Math and Applied M ath & Statistics will teach some strategies to make the most of the winter after fall classes end.

\nLearn how to best kick off or revamp your job or internship search!

\n\n

***Food will be served\
, grab it 15 minutes before the event!**

RSVP on Handshake @ https://app.joinha ndshake.com/events/108603

\n\n

Walk-ins welcome

DTSTART;TZID=America/New_York:20171205T180000 DTEND;TZID=America/New_York:20171205T193000 SEQUENCE:0 SUMMARY:Winter Wonderland: Strategies for Your Winter Job or Internship Sea rch! URL:https://engineering.jhu.edu/ams/events/winter-wonderland-strategies-win ter-job-internship-search/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10444@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:TBA

DTSTART;TZID=America/New_York:20171206T150000 DTEND;TZID=America/New_York:20171206T160000 SEQUENCE:0 SUMMARY:Data Science Seminar: Hau-Tieng Wu (Duke University) @ Hodson 203 URL:https://engineering.jhu.edu/ams/events/data-science-seminar-hau-tieng-w u-duke-university-hodson-203/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10556@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:**Title: **The Growing Importance of Satellite
Data for Health and Air Quality Applications

\n

**Abs
tract:**

Satellite data are growing in importance for healt h and air quality end users in the U.S. and around the world. From their “Gods-eye” view\, satellites provide a level of spatial coverage unobtaina ble by surface monitoring networks. Satellite observations of various pol lutants\, such as nitrogen dioxide and sulfur dioxide\, vividly demonstrat e the steady improvement of air quality in the U.S. over the last several decades thanks to environmental regulations\, such as the Clean Air Act. H owever\, while better\, U.S. air quality is still not at healthy levels an d there are occasionally extreme events (e.g.\, wildfires\, toxic spills i n Houston after Hurricane Harvey) that expose Americans to high levels of pollution. Satellite data also show that air quality in many parts of the world is rapidly degrading\, and is likely to continue to do so as the gl obal population is expected to increase by 2 billion by 2050. In this pres entation\, I will discuss the strengths and limitations of current satelli te data for health and air quality applications as well as the potential u pcoming satellites offer. I will present examples of successful uses of sa tellite data\, discuss potential uses\, and highlight ongoing challenges ( e.g.\, data processing and visualization) for satellite data end users.

\n\n

*Biographical Sketch*

Dr. Bryan Duncan is an Earth scientist at NASA’s Goddard Space Flight Center a nd has a keen interest in using NASA satellite data for societal benefit\, including for health and air quality applications. He frequently speaks t o representatives of various U.S. and international agencies (e.g.\, World Bank\, UNICEF) about how satellite data may benefit their objectives and is a member of the NASA Health and Air Quality Applied Sciences Team (HAQA ST). He is also the Project Scientist of the NASA Aura satellite mission\, which has observing air quality from space as one of its objectives.

DTSTART;TZID=America/New_York:20171207T133000 DTEND;TZID=America/New_York:20171207T143000 SEQUENCE:0 SUMMARY:The John C. & Susan S.G. Wierman Lecture Series: Bryan Duncan (NASA Goddard Space Flight Center) @ Olin 305 URL:https://engineering.jhu.edu/ams/events/john-c-susan-s-g-wierman-lecture -series-bryan-duncan-nasa-goddard-space-flight-center-tba/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10963@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Title: Principled non-convex optimization for deep learning and phase retrieval

\nAbstract: This talk looks at two classes of no n-convex problems. First\, we discuss phase retrieval problems\, and prese nt a new formulation\, called PhaseMax\, that reduces this class of non-co nvex problems into a convex linear program. Then\, we turn our attention t o more complex non-convex problems that arise in deep learning. We’ll expl ore the non-convex structure of deep networks using a range of visualizati on methods. Finally\, we discuss a class of principled algorithms for trai ning “binarized” neural networks\, and show that these algorithms have the oretical properties that enable them to overcome the non-convexities prese nt in neural loss functions.

DTSTART;TZID=America/New_York:20180131T150000 DTEND;TZID=America/New_York:20180131T160000 SEQUENCE:0 SUMMARY:Data Science Seminar: Tom Goldsten (University of Maryland College Park) @ Gilman 219 URL:https://engineering.jhu.edu/ams/events/data-science-seminar-tom-goldste n-university-of-maryland-college-park-gilman-219/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10861@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:**Title**: Approximating Minimal Cut-Generating
Functions by Extreme Functions

**Abstract**:

\n
With applications in scheduling\, networks\, and generalized assignment pr
oblems\, integer programs are ubiquitous in a variety of engineering disci
plines. Often\, integer programming algorithms make use of strategically
chosen cutting planes in order to trim the region bounded by the linear co
nstraints without removing any feasible points. Recently\, there has been
a resurgence of interest in the theory of (minimal) cut generating functio
ns\, as such functions can be used to produce quality cuts. Moreover\, th
e family of minimal functions forms a convex set\; in order to better unde
rstand this class of functions\, we wish to study the extreme functions of
this set. In this talk\, we shall see that the set of continuous minimal
cut generating functions contains a dense subset of extreme function.

**Title: ** Limit theorems for eigenvectors of
the normalized Laplacian for random graphs

\n

**Abstr
act:**

“We prove a central limit theorem for the components of the eigenvectors corresponding to the d largest eigenvalues of the nor malized Laplacian matrix of a finite dimensional random dot product graph. As a corollary\, we show that for stochastic blockmodel graphs\, the rows of the spectral embedding of the normalized Laplacian converge to multiva riate normals and furthermore the mean and the covariance matrix of each r ow are functions of the associated vertex’s block membership. Together wit h prior results for the eigenvectors of the adjacency matrix\, we then com pare\, via the Chernoff information between multivariate normal distributi ons\, how the choice of embedding method impacts subsequent inference. We demonstrate that neither embedding method dominates with respect to the in ference task of recovering the latent block assignments.”

DTSTART;TZID=America/New_York:20180208T133000 DTEND;TZID=America/New_York:20180208T143000 SEQUENCE:0 SUMMARY:AMS Seminar: Minh Hai Tang (JHU) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/ams-seminar-tba-whitehead-304-2/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10997@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:**Title: **Monotonicity of optimal contracts w
ithout the first-order approach

\n

**Abstract:**

We develop a simple sufficient condition for an optimal contrac t of a moral

\nhazard problem to be monotone in the output signal. E xisting results on monotonicity

\nrequire conditions on the output d istribution (namely\, the monotone likelihood ratio

\nproperty (MLRP )) and additional conditions to guarantee that agent’s decision is

\napproachable via the first-order approach of replacing that

\nprobl em with its first-order conditions. We know of no positive monotonicity

\nresults in the setting where the first-order approach does not apply . Indeed\, it is

\nwell-documented that when there are finitely-many possible outputs\, and the

\nfirst-order approach does not apply\, the MLRP alone is insufficient to guarantee monotonicity.

\nHowever\ , we show that when there is an interval of possible output signals\,

\nthe MLRP does suffice to establish monotonicity under additional tech nical assumptions

\nthat do not guarantee the validity of the first- order approach.

\n\n

This is joint work with Rongzhu Ke (Hong Kong Baptist University).

DTSTART;TZID=America/New_York:20180209T090000 DTEND;TZID=America/New_York:20180209T100000 SEQUENCE:0 SUMMARY:Optimization Seminar: Christopher Ryan (University of Chicago) @ Wh itehead 304 URL:https://engineering.jhu.edu/ams/events/optimization-seminar-christopher -ryan-university-chicago-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10863@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:**Title:** Maximum Likelihood Density Estimatio
n under Total Positivity

\n** **

\n**Abstract
:** Nonparametric density estimation is a challenging problem in th
eoretical statistics—in general the maximum likelihood estimate (MLE) does
not even exist! Introducing shape constraints allows a path forward. This
talk offers an invitation to non-parametric density estimation under tota
l positivity (i.e. log-supermodularity) and log-concavity. Totally positiv
e random variables are ubiquitous in real world data and possess appealing
mathematical properties. Given i.i.d. samples from such a distribution\,
we prove that the maximum likelihood estimator under these shape constrain
ts exists with probability one. We characterize the domain of the MLE and
show that it is in general larger than the convex hull of the observations
. If the observations are 2-dimensional or binary\, we show that the logar
ithm of the MLE is a tent function (i.e. a piecewise linear function) with
“poles” at the observations\, and we show that a certain convex program c
an find it. In the general case the MLE is more complicated. We give neces
sary and sufficient conditions for a tent function to be concave and super
modular\, which characterizes all the possible candidates for the MLE in t
he general case.

DTSTART;TZID=America/New_York:20180215T133000 DTEND;TZID=America/New_York:20180215T143000 SEQUENCE:0 SUMMARY:AMS Seminar: Dr. Elina Robeva (MIT) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/ams-seminar-tba-whitehead-304-3/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-11000@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:

**Title:** The Learning Premium

**Abstract**: We find equilibrium stock prices and interest rates i
n a

representative-agent model with uncertain dividends’ growth\, gradually

\nrevealed by dividends themselves\, where asset prices ar e rational –

\nreflect current information and anticipate the impact of future

\nknowledge on future prices. In addition to the usual pr emium for risk\,

\nstock returns include a learning premium\, which reflects the expected

\nchange in prices from new information. In th e long run\, the learning

\npremium vanishes\, as prices and interes t rates converge to their

\ncounterparts in the standard setting wit h known growth. The model

\nexplains the increase in price-dividend ratios of the past century if

\nboth relative risk aversion and elas ticity of intertemporal

\nsubstitution are above one. This is a join t work with Paolo Guasoni.

DTSTART;TZID=America/New_York:20180220T133000 DTEND;TZID=America/New_York:20180220T143000 SEQUENCE:0 SUMMARY:Financial Math Seminar: Maxim Bichuch (JHU) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/financial-math-seminar-maxim-bic huch-jhu-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-11096@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:Title: Data-driven modeling of vector fields and differentia l forms by spectral exterior calculus

\nAbstract: We discuss a data- driven framework for exterior calculus on manifolds. This framework is bas ed on a representations of vector fields\, differential forms\, and operat ors acting on these objects in frames (overcomplete bases) for L^2 and hig her-order Sobolev spaces built entirely from the eigenvalues and eigenfunc tions of the Laplacian of functions. Using this approach\, we represent ve ctor fields either as linear combinations of frame elements\, or as operat ors on functions via matrices. In addition\, we construct a Galerkin appro ximation scheme for the eigenvalue problem for the Laplace-de-Rham operato r on 1-forms\, and establish its spectral convergence. We present applicat ions of this scheme to a variety of examples involving data sampled on smo oth manifolds and the Lorenz 63 fractal attractor. This work is in collabo ration with Tyrus Berry.

DTSTART;TZID=America/New_York:20180221T150000 DTEND;TZID=America/New_York:20180221T160000 SEQUENCE:0 SUMMARY:Data Science Seminar: Dimitris Giannakis (NYU) @ Shaffer 304 URL:https://engineering.jhu.edu/ams/events/data-science-seminar-dimitris-gi annakis-shaffer-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10630@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:**Title:** Information\, Computation\, Optimiza
tion: Connecting the dots in the Traveling Salesman Problem

** Abstract:** Few math models scream impossible as loudly as the tr
aveling salesman problem.

\nGiven n cities\, the TSP asks for the sho rtest route to take you to all of them.

\nEasy to state\, but if P ≠ NP then no solution method can have good asymptotic

\nperformance as n goes off to infinity. The popular interpretation is that we simply

\ncannot solve realistic examples. But this skips over nearly 70 years of intense

\nmathematical study. Indeed\, in 1949 Julia Robinson describ ed the TSP challenge in

\npractical terms: “Since there are only a fi nite number of paths to consider\, the

\nproblem consists in finding a method for picking out the optimal path when n is

\nmoderately larg e\, say n = 50.” She went on to propose a linear programming attack

\nthat was adopted by her RAND colleagues Dantzig\, Fulkerson\, and Johnso n several

\nyears later.

Following in the footsteps of these
giants\, we show that a certain tour of 49\,603

\nhistoric sites in t
he US is shortest possible\, measuring distance with point-to-point

\nwalking routes obtained from Google Maps. Along the way\, we discuss the
history\,

\napplications\, and computation of this fascinating probl
em.

\n

*Biographical Sketch*

was elected a SIAM Fellow in 2009\, an INFORMS Fel low in 2010\, a member of the National Academy of Engineering in 2011\, an d an American

\nMathemat ics Society Fellow in 2012. He is the author of the popular book In Pursu it of the Traveling Salesman: Mathematics at the Limits of Computation.

\nBill is a former Editor- in-Chief of the journals Mathematical Programming (Series A and B) and Mat hematical Programming Computation. He is the past chair and current vice-c hair of the Mathematical Optimization Society and a past chair of the INFO RMS Computing Society.

DTSTART;TZID=America/New_York:20180222T133000 DTEND;TZID=America/New_York:20180222T143000 SEQUENCE:0 SUMMARY:The Goldman Distinguished Lecture Series: William Cook (University of Waterloo) @ Shaffer 100 URL:https://engineering.jhu.edu/ams/events/ams-seminar-william-cook-univers ity-waterloo-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10858@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:** Title: **PetuumMed: algorithm and system fo
r EHR-based medical decision-making

\n

**Abstract:**

With the rapid growth of electronic health records (EHRs) a nd the advancement of machine learning technologies\, needs for AI-enabled clinical decision-making support is emerging. In this talk\, I will prese nt some recent work toward these needs at Petuum Inc. where an integrative system that distills insights from large-scale and heterogeneous patient data\, as well as learns and integrates medical knowledge from broader sou rces such as the literatures and domain experts\, and empowers medical pro fessionals to make accurate and efficient decisions within the clinical fl ow\, is being built. I will discuss several aspects of practical clinical decision-support\, such as real-time information extraction from clinical notes and images\, diagnosis and treatment recommendation\, automatic repo rt generation and ICD code filling\; and the algorithmic and computational challenges behind production-quality solution to these problems.

DTSTART;TZID=America/New_York:20180301T133000 DTEND;TZID=America/New_York:20180301T143000 SEQUENCE:0 SUMMARY:AMS Seminar: Eric Xing (Carnegie Mellon) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/ams-seminar-eric-xing-carnegie-m ellon-whitehead-304/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-11131@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:**Title:** Variance swap

Title: Comparing relaxations via volume for nonconvex optimi zation

\nAbstract: Practical exact methods for global optimization o f mixed-integer nonlinear optimization formulations rely on convex relaxat ion. Then\, one way or another (via refinement and/or disjunction)\, globa l optimality is sought. Success of this paradigm depends on balancing tigh tness and lightness of relaxations. We will investigate this from a mathem atical viewpoint\, comparing polyhedral relaxations via their volumes. Spe cifically\, I will present some results concerning: fixed charge problems\ , vertex packing in graphs\, boolean quadratic formulations\, and convexif ication of monomials in the context of spatial branch-and-bound” for facto rable formulations. Our results can be employed by users (at the modeling level) and by algorithm designers/implementers alike.

DTSTART;TZID=America/New_York:20180308T133000 DTEND;TZID=America/New_York:20180308T143000 SEQUENCE:0 SUMMARY:AMS Seminar: Jon Lee (University of Michigan) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/ams-seminar-tba-whitehead-304-4/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-11106@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:TBA

DTSTART;TZID=America/New_York:20180314T150000 DTEND;TZID=America/New_York:20180314T160000 SEQUENCE:0 SUMMARY:Data Science Seminar: Edriss Titi (Texas A&M University) @ Shaffer 304 URL:https://engineering.jhu.edu/ams/events/data-science-seminar-edriss-titi -texas-am-university/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10867@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:**Title: **Bayesian monotone regression: Rates\
, coverage and tests

**Abstract:**

Shape rest rictions such as monotonicity often naturally arise. In this talk we consi der a Bayesian approach to monotone nonparametric regression with normal e rror. We assign a prior through piecewise constant functions and impose a conjugate normal prior on the coefficient. Since the resulting functions n eed not be monotone\, we project samples from the posterior on the allowed parameter space to construct a “projection posterior”. We first obtain co ntraction rates of the projection posterior distributions under various se ttings. We next obtain the limit posterior distribution of a suitably cent ered and scaled posterior distribution for the function value at a point. The limit distribution has some interesting similarity and difference with the corresponding limit distribution for the maximum likelihood estimator . By comparing the quantiles of these two distributions\, we observe an in teresting new phenomenon that coverage of a credible interval may be more than the credibility level\, an exact opposite of a phenomenon observed by Cox for smooth regression. We describe a recalibration strategy to modify the credible interval to meet the correct level of coverage. Finally we d iscuss asymptotic properties of Bayes tests for monotonicity.

\nThis talk is based on joint work with Moumita Chakraborty\, a doctoral student at North Carolina State University.

DTSTART;TZID=America/New_York:20180315T133000 DTEND;TZID=America/New_York:20180315T143000 SEQUENCE:0 SUMMARY:AMS Seminar: Subhashis Ghoshal (North Carolina State University) @ Whitehead 304 URL:https://engineering.jhu.edu/ams/events/ams-seminar-tba-whitehead-304-5/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-11155@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:**Presentation – Career Opportunities at the NSA Targe
ting AMS\, CS\, ECE\, JHUISI **

\n

When: Wednesday\, March 28\, 2018 5:30 pm – 8:00 pm EDT

\nWhere: Hodson\, 110\, Balti more\, MD 21218\, United States

\nThe National Security Agency (NSA ) currently has opportunities for highly motivated researchers to provide expertise\, guidance and support to the development and implementation of mission capabilities that align with mission driven challenges.

\nTh e Advanced Computing Systems Research Program (ACS) at NSA is looking to h ire talented researchers for a variety of positions. The ACS mission is to collaborate with industry\, academia\, and the government to drive innova tive research that will improve advanced computing systems for a range of mission applications including cybersecurity\, cryptanalysis\, and complex data analytics. The ACS has significant research projects in neuromorphic and probabilistic computing\, novel computer architectures and technologi es\, advanced modeling and simulation\, energy efficiency\, productivity\, and resilience.

\nDr. David J. Mountain will describe opportunities at NSA\, using his 36 year career as an example. He will also provide an overview of the ACS program and highlight specific research positions curr ently available. This will be followed by an open Q&A.

\nDr. Mountai n is the Senior Technical Director at the Laboratory for Physical Sciences at Research Park\, a Department of Defense research lab in Catonsville\, MD. He received a BS in Electrical Engineering from the University of Notr e Dame in 1982\, an MS in Electrical Engineering from the University of Ma ryland\, College Park\, in 1986\, and a PhD in Computer Engineering from t he University of Maryland\, Baltimore County\, in 2017. His personal resea rch projects have included radiation effects studies\, hot carrier reliabi lity characterization\, and chip-on-flex process development utilizing ult ra-thin circuits. He has been actively involved with 3D electronics resear ch for 25 years and is presently focused on specialized architectures to s upport advanced neural networks and tensor analysis. Dr. Mountain is the a uthor of more than two dozen papers\, has been awarded eight patents\, and is a Senior Member of the IEEE.

\nStudents RSVP @ https://app.joinhandshake.com/events/ 135261

\nFaculty and Staff RSVP @ https://www.eventbrite.c om/e/career-opportunities-at-the-nsa-targeting-cs-ams-ece-jhuisi-informati on-session-tickets-43894265931?aff=affiliate1

\n

\nNote:< /p>\n

- \n
- *
**Food will be served at this even during the first 30 minutes\, Please arrive early. The program will begin promptly at 6pm!< /strong>*** - For additional information about this event\, please c ontact Dr. Antwan D. Clark at aclark66@j hu.edu. \n

DTSTART;TZID=America/New_York:20180328T173000 DTEND;TZID=America/New_York:20180328T200000 SEQUENCE:0 SUMMARY:Dr. David J. Mountain will present on Career Opportunities at the N SA in Hodson 110 URL:https://engineering.jhu.edu/ams/events/dr-david-j-mountain-will-present -on-career-opportunities-at-the-nsa-in-hodson-110/ X-COST-TYPE:free END:VEVENT BEGIN:VEVENT UID:ai1ec-10871@engineering.jhu.edu/ams DTSTAMP:20180317T041026Z CATEGORIES: CONTACT: DESCRIPTION:

**Title:** TBA

\n

**Abstra
ct:** TBA

**Title:** TBA

\n

**Abstra
ct:** TBA

**Title: **Adaptive Robust Control Under Model
Uncertainty

**Abstract: **We propose a new methodolo
gy\, called adaptive robust control\, for solving a discrete-time Markovia
n control problem subject to Knightian uncertainty. We apply the general f
ramework to a financial hedging problem where the uncertainty comes from t
he fact that the true law of the underlying model is only known to belong
to a certain family of probability laws. We provide a learning algorithm t
hat reduces the model uncertainty through progressive learning about the u
nknow system. One of the pillars in the proposed methodology is the recurs
ive construction of the confidence sets for the unknown parameter. This al
lows\, in particular\, to establish the corresponding Bellman system of eq
uations.

**Title:** TBA

\n

**Abstra
ct:** TBA

**Title:** TBA

\n

**Abstra
ct:** TBA

**Title: **TBA

\n

**Abstra
ct: **TBA

**Title:** TBA

\n

**Abstra
ct:** TBA

**Title:** TBA

\n

**Abstra
ct:** TBA

**Title:** TBA

\n

**Abstra
ct:** TBA