Characterizing the Gap Between Actor-Critic and Policy Gradient

2021-06-13 · Junfeng Wen, Saurabh Kumar, Ramki Gummadi, Dale Schuurmans

Actor-critic (AC) methods are ubiquitous in reinforcement learning. Although it is understood that AC methods are closely related to policy gradient (PG), their precise connection has not been fully characterized previously. In this paper, we explain the gap between AC and PG methods by identifying the exact adjustment to the AC objective/gradient that recovers the true policy gradient of the cumulative reward objective (PG). Furthermore, by viewing the AC method as a two-player Stackelberg game between the actor and critic, we show that the Stackelberg policy gradient can be recovered as a special case of our more general analysis. Based on these results, we develop practical algorithms, Residual Actor-Critic and Stackelberg Actor-Critic, for estimating the correction between AC and PG and use these to modify the standard AC algorithm. Experiments on popular tabular and continuous environments show the proposed corrections can improve both the sample efficiency and final performance of existing AC methods.

📄 PDF Abstract BibTeX arXiv:2106.06932

Code (0)

등록된 구현이 없습니다.

Neural Policy Gradient Methods: Global Optimality and Rates of Convergence

2019-08-29 · ICLR 2020 1 · Lingxiao Wang, Qi Cai, Zhuoran Yang, Zhaoran Wang

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 "neur…

Policy Gradient Methods

On the Sample Complexity of Actor-Critic Method for Reinforcement Learning with Function Approximation

2019-10-18 · Harshat Kumar, Alec Koppel, Alejandro Ribeiro

Reinforcement learning, mathematically described by Markov Decision Problems, may be approached either through dynamic programming or policy search. Actor-critic algorithms combine the merits of both approaches by altern…

reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)

Projected Natural Actor-Critic

2013-12-01 · NeurIPS 2013 12 · Philip S. Thomas, William C. Dabney, Stephen Giguere, Sridhar Mahadevan

Natural actor-critics are a popular class of policy search algorithms for finding locally optimal policies for Markov decision processes. In this paper we address a drawback of natural actor-critics that limits their rea…

reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)

How to Learn a Useful Critic? Model-based Action-Gradient-Estimator Policy Optimization

2020-04-29 · NeurIPS 2020 12 · Pierluca D'Oro, Wojciech Jaśkowski

Deterministic-policy actor-critic algorithms for continuous control improve the actor by plugging its actions into the critic and ascending the action-value gradient, which is obtained by chaining the actor's Jacobian ma…

continuous-control Continuous Control MuJoCo

Bayesian policy gradient and actor-critic algorithms

2026-04-30 · Mohammad Ghavamzadeh, Yaakov Engel, Michal Valko arxiv

Policy gradient methods are reinforcement learning algorithms that adapt a parameterized policy by following a performance gradient estimate. Conventional policy gradient methods use Monte-Carlo techniques to estimate th…

Reinforcement Learning