English
Related papers

Related papers: Optimism Without Regularization: Constant Regret i…

200 papers

We study online reinforcement learning in average-reward stochastic games (SGs). An SG models a two-player zero-sum game in a Markov environment, where state transitions and one-step payoffs are determined simultaneously by a learner and an…

Machine Learning · Computer Science 2017-12-05 Chen-Yu Wei , Yi-Te Hong , Chi-Jen Lu

This paper considers the stability of online learning algorithms and its implications for learnability (bounded regret). We introduce a novel quantity called {\em forward regret} that intuitively measures how good an online learning…

Machine Learning · Computer Science 2012-11-28 Ankan Saha , Prateek Jain , Ambuj Tewari

Adaptivity to the difficulties of a problem is a key property in sequential decision-making problems to broaden the applicability of algorithms. Follow-the-regularized-leader (FTRL) has recently emerged as one of the most promising…

Machine Learning · Computer Science 2024-02-14 Taira Tsuchiya , Shinji Ito , Junya Honda

We show that, for any sufficiently small fixed $\epsilon > 0$, when both players in a general-sum two-player (bimatrix) game employ optimistic mirror descent (OMD) with smooth regularization, learning rate $\eta = O(\epsilon^2)$ and $T =…

Computer Science and Game Theory · Computer Science 2022-10-10 Ioannis Anagnostides , Gabriele Farina , Ioannis Panageas , Tuomas Sandholm

Recent literature on online learning has focused on developing adaptive algorithms that take advantage of a regularity of the sequence of observations, yet retain worst-case performance guarantees. A complementary direction is to develop…

Machine Learning · Computer Science 2015-01-27 Ali Jadbabaie , Alexander Rakhlin , Shahin Shahrampour , Karthik Sridharan

Fictitious play (FP) is a canonical game-theoretic learning algorithm which has been deployed extensively in decentralized control scenarios. However standard treatments of FP, and of many other game-theoretic models, assume rather…

Optimization and Control · Mathematics 2016-09-29 Brian Swenson , Soummya Kar , João Xavier , David S. Leslie

We study the problem of \emph{dynamic regret minimization} in $K$-armed Dueling Bandits under non-stationary or time varying preferences. This is an online learning setup where the agent chooses a pair of items at each round and observes…

Machine Learning · Computer Science 2022-06-14 Aadirupa Saha , Shubham Gupta

We develop provably efficient reinforcement learning algorithms for two-player zero-sum finite-horizon Markov games with simultaneous moves. To incorporate function approximation, we consider a family of Markov games where the reward…

Machine Learning · Computer Science 2020-06-25 Qiaomin Xie , Yudong Chen , Zhaoran Wang , Zhuoran Yang

Reverse Kullback-Leibler (KL) divergence-based regularization with respect to a fixed reference policy is widely used in modern reinforcement learning to preserve the desired traits of the reference policy and sometimes to promote…

Machine Learning · Computer Science 2026-02-05 Anupam Nayak , Tong Yang , Osman Yagan , Gauri Joshi , Yuejie Chi

While fictitious play is guaranteed to converge to Nash equilibrium in certain game classes, such as two-player zero-sum games, it is not guaranteed to converge in non-zero-sum and multiplayer games. We show that fictitious play in fact…

Computer Science and Game Theory · Computer Science 2024-07-30 Sam Ganzfried

Regularized online learning is widely used in machine learning applications. In online learning, performing exact minimization ($i.e.,$ implicit update) is known to be beneficial to the numerical stability and structure of solution. In this…

Machine Learning · Computer Science 2019-02-08 Chaobing Song , Ji Liu , Han Liu , Yong Jiang , Tong Zhang

Existing studies on provably efficient algorithms for Markov games (MGs) almost exclusively build on the "optimism in the face of uncertainty" (OFU) principle. This work focuses on a different approach of posterior sampling, which is…

Machine Learning · Computer Science 2022-10-06 Wei Xiong , Han Zhong , Chengshuai Shi , Cong Shen , Tong Zhang

Fictitious play (FP) is a natural learning dynamic in two-player zero-sum games. Samuel Karlin conjectured in 1959 that FP converges at a rate of $O(t^{-1/2})$ to Nash equilibrium, where $t$ is the number of steps played. However,…

Computer Science and Game Theory · Computer Science 2025-07-15 Yuanhao Wang

Self-play via online learning is one of the premier ways to solve large-scale two-player zero-sum games, both in theory and practice. Particularly popular algorithms include optimistic multiplicative weights update (OMWU) and optimistic…

Computer Science and Game Theory · Computer Science 2025-01-22 Yang Cai , Gabriele Farina , Julien Grand-Clément , Christian Kroer , Chung-Wei Lee , Haipeng Luo , Weiqiang Zheng

Swap regret is a notion that has proven itself to be central to the study of general-sum normal-form games, with swap-regret minimization leading to convergence to the set of correlated equilibria and guaranteeing non-manipulability against…

Computer Science and Game Theory · Computer Science 2025-02-28 Eshwar Ram Arunachaleswaran , Natalie Collina , Yishay Mansour , Mehryar Mohri , Jon Schneider , Balasubramanian Sivan

In this paper we study two-player bilinear zero-sum games with constrained strategy spaces. An instance of natural occurrences of such constraints is when mixed strategies are used, which correspond to a probability simplex constraint. We…

Computer Science and Game Theory · Computer Science 2022-06-10 Andre Wibisono , Molei Tao , Georgios Piliouras

In a Stackelberg game, a leader commits to a randomized strategy, and a follower chooses their best strategy in response. We consider an extension of a standard Stackelberg game, called a discrete-time dynamic Stackelberg game, that has an…

Computer Science and Game Theory · Computer Science 2022-02-11 Niklas Lauffer , Mahsa Ghasemi , Abolfazl Hashemi , Yagiz Savas , Ufuk Topcu

We consider reinforcement learning (RL) in episodic MDPs with adversarial full-information reward feedback and unknown fixed transition kernels. We propose two model-free policy optimization algorithms, POWER and POWER++, and establish…

Machine Learning · Computer Science 2020-07-02 Yingjie Fei , Zhuoran Yang , Zhaoran Wang , Qiaomin Xie

This paper examines the convergence of no-regret learning in games with continuous action sets. For concreteness, we focus on learning via "dual averaging", a widely used class of no-regret learning schemes where players take small steps…

Optimization and Control · Mathematics 2018-01-17 Panayotis Mertikopoulos , Zhengyuan Zhou

We propose a new algorithm for adversarial multi-armed bandits with unrestricted delays. The algorithm is based on a novel hybrid regularizer applied in the Follow the Regularized Leader (FTRL) framework. It achieves…

Machine Learning · Computer Science 2020-06-17 Julian Zimmert , Yevgeny Seldin