Talk by Ali Kavis (UT Austin)

on YouTube ↗27m

" summary

Presents a parameter-free second-order optimistic method for monotone convex–concave min–max problems that removes no line search or subsolvers, proves the faster O(1/T^{1.5}) rate under a Lipschitz Jacobian assumption and a mild extra assumption (operator Lipschitz) to keep iterates bounded, and uses a Hessian-regularized Newton-style update plus an explicit adaptive step rule. Experiments show parity with line-search in low dimension and clear advantage as dimension grows; deterministic tests work, stochastic extensions remain nontrivial.

AI-written summary · about

" chapters (8)

1. 0:00Problem setting
2. 3:32Operator formulation
3. 5:45Literature recap
4. 8:58Second-order challenges
5. 11:28Optimistic method
6. 16:34Algorithm goals
7. 16:44Proposed adaptive method
8. 25:48Q&A and discussion

" related talks

• 46mSmooth Nonconvex Min-Max OptimizationMeisam Razaviyayn · Adversarial Approaches in Machine Learning 2022
• 01h 04mLast-Iterate Convergence in Constrained Min-Max Optimization: SOS to the RescueYang Cai · Adversarial Approaches in Machine Learning 2022
• 58mContinuous Methods 2Maryam Fazel, Lorenzo Orecchia, Stefanie Jegelka · Bridging Continuous and Discrete Optimization Boot Camp 2017
• 57mComputationally Efficient Alternatives to Nonconvex-Nonconcave Min-Max OptimizationNisheeth Vishnoi · Adversarial Approaches in Machine Learning 2022
• 12mEquilibrium Computation and the Foundations of Multi-Agent Machine LearningEmmanouil Zampetakis · Meet the Fellows Welcome Event
• 01h 00mMin-Max Optimization from a Dynamical Systems ViewpointPanayotis Mertikopoulos · Adversarial Approaches in Machine Learning 2022
• 01h 18mOptimization IBen Recht · Big Data Boot Camp 2013
• 01h 07mStochastic Second Order Optimization Methods IFred Roosta · Foundations of Data Science Boot Camp 2018