News

Innovative new method could make computer calculations more efficient.

Quanta Magazine features Ben Grimmer's recent study on a traditional assumption in gradient descent, revealing that breaking the rule of small steps can lead to nearly three times faster results on optimization problems.

Alumna and data scientist at the Johns Hopkins University Applied Physics Laboratory uses statistical hypothesis testing to determine whether a performance discrepancy in AI can be either the product of bias or random chance.

Associate professor, Tamás Budavarí employs "mathematical language" to investigate what could best improve the quality of life in blighted communities.

Faculty, postdoctoral fellow, and PhD candidate in the department will present their accepted publications at the 2022 Neural Information Processing Systems (NeurIPS) conference.

Women with HIV frequently experience neuropsychiatric disorders such as depression, anxiety, and even cognitive impairment. Until now, though, the disordered physiological mechanisms underlying such symptoms was unclear. A study by…

On Design Day, a team of undergraduate students determine malignant or benign breast cancer images using machine learning algorithms.

Scientists at Johns Hopkins say they have identified an average level of biological and anatomical brain changes linked to Alzheimer’s disease that occur three to 10 years — some even more than 30 years — before the disease’s first recognizable symptoms appear.

Dr. Tamás Budavári, along with Kathryn Edin and Michael Braverman are recipients of the Institute for Data-Intensive Engineering and Science seed funding award.

Faculty member Greg Eyink led a team of interdisciplinary researchers to propose that turbulence causes magnetic field misbehaviors in solar flares.

It’s question that has been puzzling neuroscientists for nearly a century: How do patterns of activity in neurons generate behavior? In other words, how does nerve activity in living things,…