中文
相关论文

相关论文: Independent Natural Policy Gradient Methods for Po…

200 篇论文

This work focuses on the entropy-regularized independent natural policy gradient (NPG) algorithm in multi-agent reinforcement learning. In this work, agents are assumed to have access to an oracle with exact policy evaluation and seek to…

机器学习 · 计算机科学 2024-05-07 Youbang Sun , Tao Liu , P. R. Kumar , Shahin Shahrampour

This work studies an independent natural policy gradient (NPG) algorithm for the multi-agent reinforcement learning problem in Markov potential games. It is shown that, under mild technical assumptions and the introduction of the…

机器学习 · 计算机科学 2023-10-30 Youbang Sun , Tao Liu , Ruida Zhou , P. R. Kumar , Shahin Shahrampour

This paper investigates the problem of computing the equilibrium of competitive games, which is often modeled as a constrained saddle-point optimization problem with probability simplex constraints. Despite recent efforts in understanding…

最优化与控制 · 数学 2023-01-23 Shicong Cen , Yuting Wei , Yuejie Chi

We examine global non-asymptotic convergence properties of policy gradient methods for multi-agent reinforcement learning (RL) problems in Markov potential games (MPG). To learn a Nash equilibrium of an MPG in which the size of state space…

机器学习 · 计算机科学 2022-08-08 Dongsheng Ding , Chen-Yu Wei , Kaiqing Zhang , Mihailo R. Jovanović

Multi-agent interactions are increasingly important in the context of reinforcement learning, and the theoretical foundations of policy gradient methods have attracted surging research interest. We investigate the global convergence of…

最优化与控制 · 数学 2023-03-21 Sarath Pattathil , Kaiqing Zhang , Asuman Ozdaglar

Natural policy gradient (NPG) methods are among the most widely used policy optimization algorithms in contemporary reinforcement learning. This class of methods is often applied in conjunction with entropy regularization -- an algorithmic…

机器学习 · 统计学 2022-09-13 Shicong Cen , Chen Cheng , Yuxin Chen , Yuting Wei , Yuejie Chi

Softmax policy gradient is a popular algorithm for policy optimization in single-agent reinforcement learning, particularly since projection is not needed for each gradient update. However, in multi-agent systems, the lack of central…

最优化与控制 · 数学 2022-11-01 Runyu Zhang , Jincheng Mei , Bo Dai , Dale Schuurmans , Na Li

Constrained Markov games offer a formal mathematical framework for modeling multi-agent reinforcement learning problems where the behavior of the agents is subject to constraints. In this work, we focus on the recently introduced class of…

机器学习 · 计算机科学 2024-02-29 Philip Jordan , Anas Barakat , Niao He

We obtain global, non-asymptotic convergence guarantees for independent learning algorithms in competitive reinforcement learning settings with two agents (i.e., zero-sum stochastic games). We consider an episodic setting where in each…

机器学习 · 计算机科学 2021-01-13 Constantinos Daskalakis , Dylan J. Foster , Noah Golowich

This letter studies multi-agent reinforcement learning in partially observable Markov potential games. Solving this problem is challenging due to partial observability, decentralized information, and the curse of dimensionality. First, to…

多智能体系统 · 计算机科学 2026-04-02 Wonseok Yang , Thinh T. Doan

We address in this paper Reinforcement Learning (RL) among agents that are grouped into teams such that there is cooperation within each team but general-sum (non-zero sum) competition across different teams. To develop an RL method that…

机器学习 · 计算机科学 2025-02-11 Muhammad Aneeq uz Zaman , Alec Koppel , Mathieu Laurière , Tamer Başar

Markov Potential Games (MPGs) form an important sub-class of Markov games, which are a common framework to model multi-agent reinforcement learning problems. In particular, MPGs include as a special case the identical-interest setting where…

机器学习 · 计算机科学 2024-08-16 Pragnya Alatur , Anas Barakat , Niao He

We study Markov potential games under the infinite horizon average reward criterion. Most previous studies have been for discounted rewards. We prove that both algorithms based on independent policy gradient and independent natural policy…

机器学习 · 计算机科学 2024-03-12 Min Cheng , Ruida Zhou , P. R. Kumar , Chao Tian

In this paper, we investigate the impact of introducing relative entropy regularization on the Nash Equilibria (NE) of General-Sum $N$-agent games, revealing the fact that the NE of such games conform to linear Gaussian policies. Moreover,…

计算机科学与博弈论 · 计算机科学 2024-09-16 Muhammad Aneeq uz Zaman , Shubham Aggarwal , Melih Bastopcu , Tamer Başar

Natural policy gradient (NPG) methods with entropy regularization achieve impressive empirical success in reinforcement learning problems with large state-action spaces. However, their convergence properties and the impact of entropy…

机器学习 · 计算机科学 2026-02-17 Semih Cayci , Niao He , R. Srikant

Multi-agent reinforcement learning has been successfully applied to fully-cooperative and fully-competitive environments, but little is currently known about mixed cooperative/competitive environments. In this paper, we focus on a…

机器学习 · 计算机科学 2021-10-22 Roy Fox , Stephen McAleer , Will Overman , Ioannis Panageas

Finding Nash equilibria in two-player zero-sum imperfect-information games remains a central challenge in multi-agent reinforcement learning. Recent multi-round regularization methods offer a promising direction, yet existing approaches…

机器学习 · 计算机科学 2026-05-01 Eason Yu , Tzu Hao Liu , Clément L. Canonne , Yunke Wang , Chang Xu , Nguyen H. Tran , Stefano V. Albrecht

Policy gradient methods have become a staple of any single-agent reinforcement learning toolbox, due to their combination of desirable properties: iterate convergence, efficient use of stochastic trajectory feedback, and theoretically-sound…

计算机科学与博弈论 · 计算机科学 2025-07-10 Mingyang Liu , Gabriele Farina , Asuman Ozdaglar

This paper studies a policy optimization problem arising from collaborative multi-agent reinforcement learning in a decentralized setting where agents communicate with their neighbors over an undirected graph to maximize the sum of their…

最优化与控制 · 数学 2022-09-07 Jinchi Chen , Jie Feng , Weiguo Gao , Ke Wei

Potential games are arguably one of the most important and widely studied classes of normal form games. They define the archetypal setting of multi-agent coordination as all agent utilities are perfectly aligned with each other via a common…

机器学习 · 计算机科学 2025-09-24 Stefanos Leonardos , Will Overman , Ioannis Panageas , Georgios Piliouras
‹ 上一页 1 2 3 10 下一页 ›