Adversarial Approaches in Machine Learning

Program

Learning and Games

Location

Calvin Lab Auditorium

Date

Tuesday, Feb. 22 – Friday, Feb. 25, 2022

Back to calendar

All talks are in Pacific Time.

8:30 – 8:55 a.m.

Coffee and Check-In

8:55 – 9 a.m.

Opening Remarks

9 – 9:45 a.m.

Last-Iterate Convergence in Constrained Min-Max Optimization: SOS to the Rescue

Yang Cai (Yale University)

9:45 – 10:15 a.m.

Break

10:15 – 11 a.m.

Optimistic Gradient Methods for Multi-Agent Learning

Kaiqing Zhang (MIT)

11 – 11:30 a.m.

Break

11:30 a.m. – 12:15 p.m.

Fast Rates for No-Regret Learning in General Games

Noah Golowich (MIT)

12:15 – 2 p.m.

Lunch

2 – 2:45 p.m.

Offline Reinforcement Learning With Only Realizability

Jason Lee (Princeton)

2:45 – 3:15 p.m.

Break

3:15 – 4 p.m.

The Variational Method of Moments

Nathan Kallus (Cornell University)

4 – 5 p.m.

Reception

8:30 – 9 a.m.

Coffee and Check-In

9 – 9:45 a.m.

Min-max Optimization: From Complexity to Algorithms

Emmanouil Zampetakis (UC Berkeley)

9:45 – 10:15 a.m.

Break

10:15 – 11 a.m.

Min-Max Optimization from a Dynamical Systems Viewpoint

Panayotis Mertikopoulos (CNRS)

11 – 11:30 a.m.

Break

11:30 a.m. – 12:15 p.m.

Smooth Nonconvex Min-Max Optimization

Meisam Razaviyayn (University of Southern California)

12:15 – 2 p.m.

Lunch

2 – 2:45 p.m.

Studying Failure Modes of Deep Learning and Ways To Fix Them

Soheil Feizi (University of Maryland, College Park)

2:45 – 3:15 p.m.

Break

3:15 – 4 p.m.

Is Minimax Optimization the Right Framework to Understand Learning in Minimax Games?

Gauthier Gidel (Université de Montréal)

4 – 4:15 p.m.

Break

4:15 – 5 p.m.

Beyond Adversarial Approaches on Distributional Robustness

Qi Lei (Princeton)

8:30 – 9 a.m.

Coffee and Check-In

9 – 9:45 a.m.

Computationally Efficient Alternatives to Nonconvex-Nonconcave Min-Max Optimization

Nisheeth Vishnoi (Yale University)

9:45 – 10:15 a.m.

Break

10:15 – 11 a.m.

Halpern Iteration and Equilibria Problems

Jelena Diakonikolas (University of Wisconsin-Madison)

11 – 11:30 a.m.

Break

11:30 a.m. – 12:15 p.m.

Zeroth-Order Methods for Convex-Concave Minmax Problems: Learning from Strategically Generated Data

Chinmay Maheshwari (UC Berkeley)

12:15 – 2 p.m.

Lunch

2 – 2:45 p.m.

Are Single-Loop Algorithms Sufficient for Unbalanced Minimax Optimization？

Niao He (ETH Zürich)

2:45 – 3:15 p.m.

Break

3:15 – 4 p.m.

Policy Gradient: Optimal Estimation, Convergence, and Generalization beyond Cumulative Rewards

Mengdi Wang (Princeton University)

4 – 4:15 p.m.

Break

4:15 – 5 p.m.

Of Moments and Matching: Trade-offs and Treatments in Imitation Learning

Steven Wu (Carnegie Mellon University)

8:30 – 9 a.m.

Coffee and Check-In

9 – 9:45 a.m.

Learning Uninformative Representations

Richard Zemel (Columbia University)

9:45 – 10:15 a.m.

Break

10:15 – 11 a.m.

Generalized Energy-Based Models

Arthur Gretton (UCL)

11 – 11:30 a.m.

Break

11:30 a.m. – 12:15 p.m.

Rebel: Combining Deep Reinforcement Learning and Search for Imperfect-Information Games

Noam Brown (Facebook AI Research)

12:15 – 2 p.m.

Lunch

2 – 2:45 p.m.

Online Adversarial Multicalibration And (Multi)Calibeating

Aaron Roth (University of Pennsylvania)

2:45 – 3:15 p.m.

Break

3:15 – 4 p.m.

Adversarial Machine Learning and Instrumental Variables for Flexible Causal Modeling

Vasilis Syrgkanis (Microsoft Research)