English
Related papers

Related papers: On the Second-Order Convergence of Biased Policy G…

200 papers

We study the convergence of the actor-critic algorithm with nonlinear function approximation under a nonconvex-nonconcave primal-dual formulation. Stochastic gradient descent ascent is applied with an adaptive proximal term for robust…

Machine Learning · Computer Science 2022-03-01 Jing Dong , Li Shen , Yinggan Xu , Baoxiang Wang

We consider large-scale Markov decision processes with an unknown cost function and address the problem of learning a policy from a finite set of expert demonstrations. We assume that the learner is not allowed to interact with the expert…

Machine Learning · Computer Science 2021-12-30 Angeliki Kamoutsi , Goran Banjac , John Lygeros

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

Off-policy learning ability is an important feature of reinforcement learning (RL) for practical applications. However, even one of the most elementary RL algorithms, temporal-difference (TD) learning, is known to suffer form divergence…

Machine Learning · Computer Science 2025-04-21 Han-Dong Lim , Donghwan Lee

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

Despite its popularity in the reinforcement learning community, a provably convergent policy gradient method for continuous space-time control problems with nonlinear state dynamics has been elusive. This paper proposes proximal gradient…

Optimization and Control · Mathematics 2022-12-27 Christoph Reisinger , Wolfgang Stockinger , Yufei Zhang

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

We consider the problem of finding an approximate second-order stationary point of a constrained non-convex optimization problem. We first show that, unlike the gradient descent method for unconstrained optimization, the vanilla projected…

Optimization and Control · Mathematics 2020-06-04 Maher Nouiehed , Jason D. Lee , Meisam Razaviyayn

In reinforcement learning, off-policy actor-critic methods like DDPG and TD3 use deterministic policy gradients: the Q-function is learned from environment data, while the actor maximizes it via gradient ascent. We observe that in complex…

Machine Learning · Computer Science 2025-10-13 Ayush Jain , Norio Kosaka , Xinhu Li , Kyung-Min Kim , Erdem Bıyık , Joseph J. Lim

Policy gradient methods in reinforcement learning update policy parameters by taking steps in the direction of an estimated gradient of policy value. In this paper, we consider the statistically efficient estimation of policy gradients from…

Machine Learning · Statistics 2020-02-21 Nathan Kallus , Masatoshi Uehara

Policy gradient methods with actor-critic schemes demonstrate tremendous empirical successes, especially when the actors and critics are parameterized by neural networks. However, it remains less clear whether such "neural" policy gradient…

Machine Learning · Computer Science 2020-06-23 Lingxiao Wang , Qi Cai , Zhuoran Yang , Zhaoran Wang

This paper considers policy search in continuous state-action reinforcement learning problems. Typically, one computes search directions using a classic expression for the policy gradient called the Policy Gradient Theorem, which decomposes…

Machine Learning · Computer Science 2020-04-13 Sujay Bhatt , Alec Koppel , Vikram Krishnamurthy

Policy-gradient methods in Reinforcement Learning(RL) are very universal and widely applied in practice but their performance suffers from the high variance of the gradient estimate. Several procedures were proposed to reduce it including…

Machine Learning · Computer Science 2022-06-16 Maxim Kaledin , Alexander Golubev , Denis Belomestny

Off-policy learning refers to the problem of learning the value function of a way of behaving, or policy, while following a different policy. Gradient-based off-policy learning algorithms, such as GTD and TDC/GQ, converge even when using…

Artificial Intelligence · Computer Science 2015-12-15 Lucas Lehnert , Doina Precup

The concept of the value-gradient is introduced and developed in the context of reinforcement learning. It is shown that by learning the value-gradients exploration or stochastic behaviour is no longer needed to find locally optimal…

Neural and Evolutionary Computing · Computer Science 2008-03-26 Michael Fairbank

Motivated by applications in risk-sensitive reinforcement learning, we study mean-variance optimization in a discounted reward Markov Decision Process (MDP). Specifically, we analyze a Temporal Difference (TD) learning algorithm with linear…

Machine Learning · Computer Science 2025-03-13 Tejaram Sangadi , L. A. Prashanth , Krishna Jagannathan

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…

Optimization and Control · Mathematics 2023-03-21 Sarath Pattathil , Kaiqing Zhang , Asuman Ozdaglar

Many popular practical reinforcement learning (RL) algorithms employ evolving reward functions-through techniques such as reward shaping, entropy regularization, or curriculum learning-yet their theoretical foundations remain…

Machine Learning · Computer Science 2025-10-15 Rui Hu , Yu Chen , Longbo Huang

We consider off-policy temporal-difference (TD) learning methods for policy evaluation in Markov decision processes with finite spaces and discounted reward criteria, and we present a collection of convergence results for several…

Machine Learning · Computer Science 2018-03-30 Huizhen Yu

Optimization of parameterized policies for reinforcement learning (RL) is an important and challenging problem in artificial intelligence. Among the most common approaches are algorithms based on gradient ascent of a score function…

Machine Learning · Computer Science 2020-06-15 Sriram Srinivasan , Marc Lanctot , Vinicius Zambaldi , Julien Perolat , Karl Tuyls , Remi Munos , Michael Bowling