Conference proceeding
No-Regret Learning in Time-Varying Zero-Sum Games
INTERNATIONAL CONFERENCE ON MACHINE LEARNING, Vol.162, pp.26772-26808
Proceedings of Machine Learning Research
01/01/2022
Abstract
Learning from repeated play in a fixed two-player zero-sum game is a classic problem in game theory and online learning. We consider a variant of this problem where the game payoff matrix changes over time, possibly in an adversarial manner. We first present three performance measures to guide the algorithmic design for this problem: 1) the well-studied individual regret, 2) an extension of duality gap, and 3) a new measure called dynamic Nash Equilibrium regret, which quantifies the cumulative difference between the player's payoff and the minimax game value. Next, we develop a single parameter-free algorithm that simultaneously enjoys favorable guarantees under all these three performance measures. These guarantees are adaptive to different nonstationarity measures of the payoff matrices and, importantly, recover the best known results when the payoff matrix is fixed. Our algorithm is based on a two-layer structure with a meta-algorithm learning over a group of black-box base-learners satisfying a certain property, along with several novel ingredients specifically designed for the time-varying game setting. Empirical results further validate the effectiveness of our algorithm.
Details
- Title: Subtitle
- No-Regret Learning in Time-Varying Zero-Sum Games
- Creators
- Mengxiao Zhang - Univ Southern Calif, Los Angeles, CA 90089 USAPeng Zhao - Nanjing UniversityHaipeng Luo - Univ Southern Calif, Los Angeles, CA 90089 USAZhi-Hua Zhou - Nanjing University
- Contributors
- K Chaudhuri (Editor)S Jegelka (Editor)L Song (Editor)C Szepesvari (Editor)G Niu (Editor)S Sabato (Editor)
- Resource Type
- Conference proceeding
- Publication Details
- INTERNATIONAL CONFERENCE ON MACHINE LEARNING, Vol.162, pp.26772-26808
- Publisher
- JMLR-JOURNAL MACHINE LEARNING RESEARCH
- Series
- Proceedings of Machine Learning Research
- ISSN
- 2640-3498
- eISSN
- 2640-3498
- Number of pages
- 37
- Grant note
- IIS-1943607 / NSF; National Science Foundation (NSF) 61921006 / National Science Foundation of China; National Natural Science Foundation of China (NSFC)
- Language
- English
- Date published
- 01/01/2022
- Academic Unit
- Business Analytics
- Record Identifier
- 9984701827202771
Metrics
1 Record Views