PhD Seminar • Artificial Intelligence — Newton-type Methods for Minimax Optimization

Thursday, September 24, 2020 4:00 pm - 4:00 pm EDT (GMT -04:00)

Please note: This PhD seminar will be given online.

Guojun Zhang, PhD candidate
David R. Cheriton School of Computer Science

Supervisors: Professors Yaoliang Yu and Pascal Poupart

Differential games, in particular two-player sequential games (a.k.a. minimax optimization), have been an important modeling tool in applied science and received renewed interest in machine learning due to many recent applications. To account for the sequential and nonconvex nature, new solution concepts and algorithms have been developed.

In this work, we provide a detailed analysis of existing algorithms and relate them to two novel Newton-type algorithms. We argue that our Newton-type algorithms nicely complement existing ones in that (a) they converge faster to (strict) local minimax points; (b) they are much more effective when the problem is ill-conditioned; (c) their computational complexity remains similar. We verify our theoretical results by conducting experiments on training GANs.


To join this PhD seminar on Zoom, please go to https://vectorinstitute.zoom.us/j/98670343311?pwd=bDUweHQ5NTl1NmFvNmd3cVNHNEM3Zz09.