English
Related papers

Related papers: Moderate Actor-Critic Methods: Controlling Overest…

200 papers

We propose the first model-free algorithm that achieves low regret performance for decentralized learning in two-player zero-sum tabular stochastic games with infinite-horizon average-reward objective. In decentralized learning, the…

Machine Learning · Computer Science 2023-01-16 Romain Cravic , Nicolas Gast , Bruno Gaujal

Double Q-learning is a popular reinforcement learning algorithm in Markov decision process (MDP) problems. Clipped Double Q-learning, as an effective variant of Double Q-learning, employs the clipped double estimator to approximate the…

Machine Learning · Computer Science 2021-05-04 Haobo Jiang , Jin Xie , Jian Yang

The overestimation bias is one of the major impediments to accurate off-policy learning. This paper investigates a novel way to alleviate the overestimation bias in a continuous control setting. Our method---Truncated Quantile Critics,…

Machine Learning · Computer Science 2020-05-12 Arsenii Kuznetsov , Pavel Shvechikov , Alexander Grishin , Dmitry Vetrov

Actor-critic algorithms have become a cornerstone in reinforcement learning (RL), leveraging the strengths of both policy-based and value-based methods. Despite recent progress in understanding their statistical efficiency, no existing work…

Machine Learning · Statistics 2025-05-07 Kevin Tan , Wei Fan , Yuting Wei

The Mean Square Error (MSE) is commonly utilized to estimate the solution of the optimal value function in the vast majority of offline reinforcement learning (RL) models and has achieved outstanding performance. However, we find that its…

Machine Learning · Computer Science 2024-06-06 Yu Zhang , Rui Yu , Zhipeng Yao , Wenyuan Zhang , Jun Wang , Liming Zhang

This paper proposes a reinforcement learning (RL)-based backstepping control strategy to achieve fixed time consensus in nonlinear multi-agent systems with strict feedback dynamics. Agents exchange only output information with their…

Systems and Control · Electrical Eng. & Systems 2025-07-23 Aria Delshad , Maryam Babazadeh

Improving the sample efficiency of reinforcement learning algorithms requires effective exploration. Following the principle of $\textit{optimism in the face of uncertainty}$ (OFU), we train a separate exploration policy to maximize the…

Machine Learning · Computer Science 2022-11-23 Jiachen Li , Shuo Cheng , Zhenyu Liao , Huayan Wang , William Yang Wang , Qinxun Bai

Value-based algorithms are a cornerstone of off-policy reinforcement learning due to their simplicity and training stability. However, their use has traditionally been restricted to discrete action spaces, as they rely on estimating…

Machine Learning · Computer Science 2025-10-23 Yigit Korkmaz , Urvi Bhuwania , Ayush Jain , Erdem Bıyık

Multi-task reinforcement learning (RL) aims to find a single policy that effectively solves multiple tasks at the same time. This paper presents a constrained formulation for multi-task RL where the goal is to maximize the average…

Optimization and Control · Mathematics 2024-05-07 Sihan Zeng , Thinh T. Doan , Justin Romberg

This paper studies the discrete-time linear-quadratic-Gaussian mean field (MF) social control problem in an infinite horizon, where the dynamics of all agents are unknown. The objective is to design a reinforcement learning (RL) algorithm…

Optimization and Control · Mathematics 2025-12-05 Hanfang Zhang , Bing-Chang Wang , Shuo Chen

Temporal difference (TD) learning is a fundamental technique in reinforcement learning that updates value estimates for states or state-action pairs using a TD target. This target represents an improved estimate of the true value by…

Machine Learning · Computer Science 2024-08-05 Wuhao Wang , Zhiyong Chen , Lepeng Zhang

In this paper, we propose a new solution to reward adaptation (RA) in reinforcement learning, where the agent adapts to a target reward function based on one or more existing source behaviors learned a priori under the same domain dynamics…

Machine Learning · Computer Science 2025-10-23 Kevin Vora , Yu Zhang

Actor-critic methods are widely used in offline reinforcement learning practice, but are not so well-understood theoretically. We propose a new offline actor-critic algorithm that naturally incorporates the pessimism principle, leading to…

Machine Learning · Computer Science 2021-08-20 Andrea Zanette , Martin J. Wainwright , Emma Brunskill

Reinforcement learning (RL) has shown remarkable success in solving complex decision-making and control tasks. However, many model-free RL algorithms experience performance degradation due to inaccurate value estimation, particularly the…

Machine Learning · Computer Science 2025-08-07 Jingliang Duan , Wenxuan Wang , Liming Xiao , Jiaxin Gao , Shengbo Eben Li , Chang Liu , Ya-Qin Zhang , Bo Cheng , Keqiang Li

This paper investigates trajectory tracking problem for a class of underactuated autonomous underwater vehicles (AUVs) with unknown dynamics and constrained inputs. Different from existing policy gradient methods which employ single…

Machine Learning · Computer Science 2019-09-10 Wenjie Shi , Shiji Song , Cheng Wu , C. L. Philip Chen

The overestimation phenomenon caused by function approximation is a well-known issue in value-based reinforcement learning algorithms such as deep Q-networks and DDPG, which could lead to suboptimal policies. To address this issue, TD3…

Machine Learning · Computer Science 2023-11-07 Qiang He , Xinwen Hou

Traditional reinforcement learning methods optimize agents without considering safety, potentially resulting in unintended consequences. In this paper, we propose an optimal actor-free policy that optimizes a risk-sensitive criterion based…

Machine Learning · Computer Science 2023-07-04 Ruoqi Zhang , Jens Sjölund

Majority of off-policy reinforcement learning algorithms use overestimation bias control techniques. Most of these techniques rooted in heuristics, primarily addressing the consequences of overestimation rather than its fundamental origins.…

Machine Learning · Computer Science 2023-09-27 Arsenii Kuznetsov

This paper proposes a safety modulator actor-critic (SMAC) method to address safety constraint and overestimation mitigation in model-free safe reinforcement learning (RL). A safety modulator is developed to satisfy safety constraints by…

Artificial Intelligence · Computer Science 2024-10-10 Qihan Qi , Xinsong Yang , Gang Xia , Daniel W. C. Ho , Pengyang Tang

Offline reinforcement learning (RL) seeks to learn optimal policies from static datasets without further environment interaction. A key challenge is the distribution shift between the learned and behavior policies, leading to…

Machine Learning · Computer Science 2025-08-11 Haohui Chen , Zhiyong Chen