Please note: This seminar will take place virtually over Zoom.
Pavel Izmailov, PhD candidate
Computer Science Department, New York University
Please note: This seminar will take place in DC 2585.
Felix Dangel, Postdoctoral Researcher
Vector Institute for Artificial Intelligence
Popular deep learning frameworks prioritize computing the average mini-batch gradient. Yet, other quantities such as its variance or many approximations to the Hessian can be computed efficiently, and at the same time as the gradient mean. They are of great interest to researchers and practitioners, but implementing them is often burdensome or inefficient.
Please note: This master’s thesis presentation will take place online.
Anupa Murali, Master’s candidate
David R. Cheriton School of Computer Science
Supervisor: Professor Bin Ma