中文
相关论文

相关论文: Settling the Variance of Multi-Agent Policy Gradie…

200 篇论文

Recent renewed interest in multi-agent reinforcement learning (MARL) has generated an impressive array of techniques that leverage deep reinforcement learning, primarily actor-critic architectures, and can be applied to a limited range of…

机器学习 · 计算机科学 2021-06-21 Keyang He , Prashant Doshi , Bikramjit Banerjee

Multi-agent reinforcement learning systems aim to provide interacting agents with the ability to collaboratively learn and adapt to the behaviour of other agents. In many real-world applications, the agents can only acquire a partial view…

机器学习 · 计算机科学 2018-12-04 Ozsel Kilinc , Giovanni Montana

In multi-agent reinforcement learning, the behaviors that agents learn in a single Markov Game (MG) are typically confined to the given agent number. Every single MG induced by varying the population may possess distinct optimal joint…

机器学习 · 计算机科学 2023-06-06 Shenao Zhang , Li Shen , Lei Han , Li Shen

Cooperative multi-agent policy gradient (MAPG) algorithms have recently attracted wide attention and are regarded as a general scheme for the multi-agent system. Credit assignment plays an important role in MAPG and can induce cooperation…

机器学习 · 计算机科学 2023-03-07 Wubing Chen , Wenbin Li , Xiao Liu , Shangdong Yang , Yang Gao

Dynamic pricing in competitive retail markets requires strategies that adapt to fluctuating demand and competitor behavior. In this work, we present a systematic empirical evaluation of multi-agent reinforcement learning (MARL)…

机器学习 · 计算机科学 2026-03-19 Krishna Kumar Neelakanta Pillai Santha Kumari Amma

Multi-agent systems provide a powerful way to extend large language models (LLMs) by decomposing a complex task into specialized subtasks handled by different agents. However, their performance is often hindered by error propagation,…

机器学习 · 计算机科学 2026-05-14 Zheng Wang , Yuang Liu , Yangkai Ding

Policy gradient methods have become a standard for training reinforcement learning agents in a scalable and efficient manner. However, they do not account for transition uncertainty, whereas learning robust policies can be computationally…

机器学习 · 计算机科学 2023-12-12 Navdeep Kumar , Esther Derman , Matthieu Geist , Kfir Levy , Shie Mannor

Various methods for Multi-Agent Reinforcement Learning (MARL) have been developed with the assumption that agents' policies are based on accurate state information. However, policies learned through Deep Reinforcement Learning (DRL) are…

人工智能 · 计算机科学 2024-04-15 Songyang Han , Sanbao Su , Sihong He , Shuo Han , Haizhao Yang , Shaofeng Zou , Fei Miao

Multi-agent reinforcement learning (MARL) addresses sequential decision-making problems with multiple agents, where each agent optimizes its own objective. In many real-world instances, the agents may not only want to optimize their…

机器学习 · 计算机科学 2023-06-14 Pragnya Alatur , Giorgia Ramponi , Niao He , Andreas Krause

We develop a multi-agent reinforcement learning (MARL) algorithm to minimize the total energy consumption of multiple massive MIMO (multiple-input multiple-output) base stations (BSs) in a multi-cell network while preserving the overall…

信息论 · 计算机科学 2024-02-06 Tianzhang Cai , Qichen Wang , Shuai Zhang , Özlem Tuğfe Demir , Cicek Cavdar

We propose a metalearning approach for learning gradient-based reinforcement learning (RL) algorithms. The idea is to evolve a differentiable loss function, such that an agent, which optimizes its policy to minimize this loss, will achieve…

机器学习 · 计算机科学 2018-05-01 Rein Houthooft , Richard Y. Chen , Phillip Isola , Bradly C. Stadie , Filip Wolski , Jonathan Ho , Pieter Abbeel

Policy gradient methods are a widely used class of model-free reinforcement learning algorithms where a state-dependent baseline is used to reduce gradient estimator variance. Several recent papers extend the baseline to depend on both the…

机器学习 · 计算机科学 2018-11-20 George Tucker , Surya Bhupatiraju , Shixiang Gu , Richard E. Turner , Zoubin Ghahramani , Sergey Levine

Intraday surgical scheduling is a multi-objective decision problem under uncertainty-balancing elective throughput, urgent and emergency demand, delays, sequence-dependent setups, and overtime. We formulate the problem as a cooperative…

机器学习 · 计算机科学 2025-12-05 Kailiang Liu , Ying Chen , Ralf Borndörfer , Thorsten Koch

Many recent successful off-policy multi-agent reinforcement learning (MARL) algorithms for cooperative partially observable environments focus on finding factorized value functions, leading to convoluted network structures. Building on the…

机器学习 · 计算机科学 2023-10-27 Raphaël Avalos , Mathieu Reymond , Ann Nowé , Diederik M. Roijers

Multi-objective reinforcement learning (MORL) algorithms tackle sequential decision problems where agents may have different preferences over (possibly conflicting) reward functions. Such algorithms often learn a set of policies (each…

机器学习 · 计算机科学 2023-08-16 Lucas N. Alegre , Ana L. C. Bazzan , Diederik M. Roijers , Ann Nowé , Bruno C. da Silva

Multi-agent LLM systems enable advanced reasoning and tool use via role specialization, yet reliable reinforcement learning (RL) post-training for such systems remains difficult. In this work, we theoretically pinpoint a key reason for…

机器学习 · 计算机科学 2026-02-10 Lang Feng , Longtao Zheng , Shuo He , Fuxiang Zhang , Bo An

Decentralized execution is one core demand in multi-agent reinforcement learning (MARL). Recently, most popular MARL algorithms have adopted decentralized policies to enable decentralized execution, and use gradient descent as the…

多智能体系统 · 计算机科学 2025-12-30 Jianing Ye , Chenghao Li , Yongqiang Dou , Jianhao Wang , Guangwen Yang , Chongjie Zhang

In order to model risk aversion in reinforcement learning, an emerging line of research adapts familiar algorithms to optimize coherent risk functionals, a class that includes conditional value-at-risk (CVaR). Because optimizing the…

机器学习 · 计算机科学 2021-03-09 Audrey Huang , Liu Leqi , Zachary C. Lipton , Kamyar Azizzadenesheli

Adaptive cooperation in multi-agent reinforcement learning (MARL) requires policies to express homogeneous, specialised, or mixed behaviours, yet achieving this adaptivity remains a critical challenge. While parameter sharing (PS) is…

机器学习 · 计算机科学 2025-10-30 Kale-ab Abebe Tessera , Arrasy Rahman , Amos Storkey , Stefano V. Albrecht

We introduce Phasic Policy Gradient (PPG), a reinforcement learning framework which modifies traditional on-policy actor-critic methods by separating policy and value function training into distinct phases. In prior methods, one must choose…

机器学习 · 计算机科学 2020-09-10 Karl Cobbe , Jacob Hilton , Oleg Klimov , John Schulman