English
Related papers

Related papers: Second-Order Actor-Critic Methods for Discounted M…

200 papers

Variance-reduced gradient estimators for policy gradient methods have been one of the main focus of research in the reinforcement learning in recent years as they allow acceleration of the estimation process. We propose a variance-reduced…

Machine Learning · Computer Science 2023-11-28 Saber Salehkaleybar , Sadegh Khorasani , Negar Kiyavash , Niao He , Patrick Thiran

In value-based deep reinforcement learning methods, approximation of value functions induces overestimation bias and leads to suboptimal policies. We show that in deep actor-critic methods that aim to overcome the overestimation bias, if…

Machine Learning · Computer Science 2021-12-28 Baturay Saglam , Enes Duran , Dogan C. Cicek , Furkan B. Mutlu , Suleyman S. Kozat

In this work, we provide theoretical guarantees for reward decomposition in deterministic MDPs. Reward decomposition is a special case of Hierarchical Reinforcement Learning, that allows one to learn many policies in parallel and combine…

Machine Learning · Computer Science 2018-03-14 Tom Zahavy , Avinatan Hasidim , Haim Kaplan , Yishay Mansour

In this paper, we propose several new stochastic second-order algorithms for policy optimization that only require gradient and Hessian-vector product in each iteration, making them computationally efficient and comparable to policy…

Optimization and Control · Mathematics 2023-01-31 Jinsong Liu , Chenghan Xie , Qi Deng , Dongdong Ge , Yinyu Ye

The Hessian-vector product has been utilized to find a second-order stationary solution with strong complexity guarantee (e.g., almost linear time complexity in the problem's dimensionality). In this paper, we propose to further reduce the…

Optimization and Control · Mathematics 2017-10-03 Mingrui Liu , Tianbao Yang

The hierarchical interaction between the actor and critic in actor-critic based reinforcement learning algorithms naturally lends itself to a game-theoretic interpretation. We adopt this viewpoint and model the actor and critic interaction…

Machine Learning · Computer Science 2021-09-28 Liyuan Zheng , Tanner Fiez , Zane Alumbaugh , Benjamin Chasnov , Lillian J. Ratliff

We consider the problem of control in the setting of reinforcement learning (RL), where model information is not available. Policy gradient algorithms are a popular solution approach for this problem and are usually shown to converge to a…

Machine Learning · Computer Science 2023-04-24 Mizhaan Prajit Maniyar , Akash Mondal , Prashanth L. A. , Shalabh Bhatnagar

Deep Reinforcement Learning (DRL) has been extensively used to address portfolio optimization problems. The DRL agents acquire knowledge and make decisions through unsupervised interactions with their environment without requiring explicit…

Machine Learning · Computer Science 2025-01-14 Ruoyu Sun , Yue Xi , Angelos Stefanidis , Zhengyong Jiang , Jionglong Su

In traditional reinforcement learning, an agent maximizes the reward collected during its interaction with the environment by approximating the optimal policy through the estimation of value functions. Typically, given a state s and action…

Machine Learning · Computer Science 2018-06-20 Shangda Li , Selina Bing , Steven Yang

Deep reinforcement learning has obtained significant breakthroughs in recent years. Most methods in deep-RL achieve good results via the maximization of the reward signal provided by the environment, typically in the form of discounted…

Machine Learning · Computer Science 2018-09-10 Yubin Deng , Ke Yu , Dahua Lin , Xiaoou Tang , Chen Change Loy

The average reward criterion is relatively less studied as most existing works in the Reinforcement Learning literature consider the discounted reward criterion. There are few recent works that present on-policy average reward actor-critic…

Machine Learning · Computer Science 2023-07-20 Naman Saxena , Subhojyoti Khastigir , Shishir Kolathaya , Shalabh Bhatnagar

Two-time-scale optimization is a framework introduced in Zeng et al. (2024) that abstracts a range of policy evaluation and policy optimization problems in reinforcement learning (RL). Akin to bi-level optimization under a particular type…

Optimization and Control · Mathematics 2026-01-21 Sihan Zeng , Thinh T. Doan

We revisit the estimation bias in policy gradients for the discounted episodic Markov decision process (MDP) from Deep Reinforcement Learning (DRL) perspective. The objective is formulated theoretically as the expected returns discounted…

Machine Learning · Computer Science 2023-02-13 Haoxuan Pan , Deheng Ye , Xiaoming Duan , Qiang Fu , Wei Yang , Jianping He , Mingfei Sun

How to obtain good value estimation is one of the key problems in Reinforcement Learning (RL). Current value estimation methods, such as DDPG and TD3, suffer from unnecessary over- or underestimation bias. In this paper, we explore the…

Machine Learning · Computer Science 2021-06-08 Jiafei Lyu , Xiaoteng Ma , Jiangpeng Yan , Xiu Li

In value-based reinforcement learning methods such as deep Q-learning, function approximation errors are known to lead to overestimated value estimates and suboptimal policies. We show that this problem persists in an actor-critic setting…

Artificial Intelligence · Computer Science 2018-10-23 Scott Fujimoto , Herke van Hoof , David Meger

Training critiquing language models to assess and provide feedback on model outputs is a promising way to improve LLMs for complex reasoning tasks. However, existing approaches typically rely on stronger supervisors for annotating critique…

We reformulate the option framework as two parallel augmented MDPs. Under this novel formulation, all policy optimization algorithms can be used off the shelf to learn intra-option policies, option termination conditions, and a master…

Machine Learning · Computer Science 2019-09-12 Shangtong Zhang , Shimon Whiteson

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

We develop a new algorithm for non-convex stochastic optimization that finds an $\epsilon$-critical point in the optimal $O(\epsilon^{-3})$ stochastic gradient and Hessian-vector product computations. Our algorithm uses Hessian-vector…

Machine Learning · Computer Science 2021-07-13 Hoang Tran , Ashok Cutkosky

On error of value function inevitably causes an overestimation phenomenon and has a negative impact on the convergence of the algorithms. To mitigate the negative effects of the approximation error, we propose Error Controlled Actor-critic…

Machine Learning · Computer Science 2021-09-08 Xingen Gao , Fei Chao , Changle Zhou , Zhen Ge , Chih-Min Lin , Longzhi Yang , Xiang Chang , Changjing Shang