Computer Science - Machine Learning

Fast Last-Iterate Convergence of Learning in Games Requires Forgetful Algorithms
Yang Cai, Gabriele Farian, Julien Grand-Clément, Christian Kroer, Chung-Wei Lee, Haipeng Luo, Weiqiang Zheng
Last updated on Jul 1, 2024
Cite arXiv
Accelerated Algorithms for Constrained Nonconvex-Nonconcave Min-Max Optimization and Comonotone Inclusion
We study constrained comonotone min-max optimization, a structured class of nonconvex-nonconcave min-max optimization problems, and …
Yang Cai, Argyris Oikonomou, Weiqiang Zheng
PDF Cite arXiv
On Tractable $\Phi$-Equilibria in Non-Concave Games
We study constrained comonotone min-max optimization, a structured class of nonconvex-nonconcave min-max optimization problems, and …
Yang Cai, Constantinos Daskalakis, Haipeng Luo, Chen-Yu Wei, Weiqiang Zheng
Last updated on Jul 5, 2024
PDF Cite arXiv
Near-Optimal Policy Optimization for Correlated Equilibrium in General-Sum Markov Games
We study policy optimization algorithms for computing correlated equilibria in multi-player general-sum Markov Games. Previous results …
Yang Cai, Haipeng Luo, Chen-Yu Wei, Weiqiang Zheng
PDF Cite arXiv
Learning Thresholds with Latent Values and Censored Feedback
In this paper, we investigate a problem of actively learning threshold in latent space, where the unknown reward $g(\gamma, v)$ …
(*) Jiahao Zhang, Tao Lin, Weiqiang Zheng, Zhe Feng, Yifeng Teng, Xiaotie Deng
PDF Cite arXiv Openreview
Last-Iterate Convergence Properties of Regret-Matching Algorithms in Games
Algorithms based on regret matching, specifically regret matching+ (RM+), and its variants are the most popular approaches for solving …
Yang Cai, Gabriele Farian, Julien Grand-Clément, Christian Kroer, Chung-Wei Lee, Haipeng Luo, Weiqiang Zheng
Last updated on Mar 14, 2024
Cite arXiv
Uncoupled and Convergent Learning in Two-Player Zero-Sum Markov Games with Bandit Feedback
We revisit the problem of learning in two-player zero-sum Markov games, focusing on developing an algorithm that is uncoupled, …
Yang Cai, Haipeng Luo, Chen-Yu Wei, Weiqiang Zheng
PDF Cite arXiv
Tight Last-Iterate Convergence of the Extragradient and the Optimistic Gradient Descent-Ascent Algorithm for Constrained Monotone Variational Inequalities
The monotone variational inequality is a central problem in mathematical programming that unifies and generalizes many important …
Yang Cai, Argyris Oikonomou, Weiqiang Zheng
PDF Cite arXiv