AMS Weekly Seminar | Aaron Sidford

Location: Gilman 50

When: November 21st at 1:30 p.m.

Title: Theoretical Advances in Efficiently Solving Markov Decision Processes

Abstract: Markov Decision Processes (MDP) are a fundamental mathematical model for reasoning about uncertainty and are foundational to reinforcement learning theory. Over the past decade, there have been substantial advances in the design and analysis of algorithms for computing approximately optimal policies in MDPs in a variety of settings. In this talk, I will survey these advances touching upon optimization tools of potential broader utility. In particular, this talk will highlight recent joint work with Ishani Karmarkar, Jiayi Wang, and Yujia Jin on this topic (arXiv:2405.12952).

Zoom link: https://wse.zoom.us/j/96299385386?pwd=HvOzDnw7A2SOQkMrbBbr5AXjqurdj0.1