Policy Gradient

Policy gradient methods are a core component of reinforcement learning, aiming to optimize policies by directly estimating the gradient of expected cumulative rewards. Current research emphasizes improving sample efficiency and addressing challenges like high-dimensional state spaces and non-convex optimization landscapes through techniques such as residual policy learning, differentiable simulation, and novel policy architectures (e.g., tree-based, low-rank matrix models). These advancements are significant for both theoretical understanding of reinforcement learning algorithms and practical applications in robotics, control systems, and other domains requiring efficient and robust decision-making under uncertainty.

Papers

June 10, 2023

A Single-Loop Deep Actor-Critic Algorithm for Constrained Reinforcement Learning with Provable Convergence
Kexuan Wang, An Liu, Baishuo Lin
Reinforcement Learning Policy Gradient Actor Critic Actor Critic Algorithm Constrained Reinforcement Learning Single Loop Provable Convergence

June 1, 2023

Achieving Fairness in Multi-Agent Markov Decision Processes Using Reinforcement Learning
Peizhong Ju, Arnob Ghosh, Ness B. Shroff
Reinforcement Learning Markov Decision Process Multi Agent Policy Gradient Achieving Fairness Balancing Fairness

May 31, 2023

May 30, 2023

May 29, 2023

DoMo-AC: Doubly Multi-step Off-policy Actor-Critic Algorithm
Yunhao Tang, Tadashi Kozuno, Mark Rowland, Anna Harutyunyan, Rémi Munos, Bernardo Ávila Pires, Michal Valko
Policy Gradient Policy Learning Policy Improvement Policy Actor Critic Multi Step Off Policy

May 26, 2023

A Policy Gradient Method for Confounded POMDPs
Mao Hong, Zhengling Qi, Yanxun Xu
Markov Decision Process Policy Gradient Partially Observable Markov Decision Process

May 24, 2023

May 23, 2023

Proximal Policy Gradient Arborescence for Quality Diversity Reinforcement Learning
Sumeet Batra, Bryon Tjanaka, Matthew C. Fontaine, Aleksei Petrenko, Stefanos Nikolaidis, Gaurav Sukhatme
Reinforcement Learning Policy Gradient Quality Diversity Diversity Reinforcement Learning Diverse Agent

May 22, 2023

Training Diffusion Models with Reinforcement Learning
Kevin Black, Michael Janner, Yilun Du, Ilya Kostrikov, Sergey Levine
Reinforcement Learning Diffusion Model Text to Image Diffusion Model Policy Gradient

May 20, 2023

May 18, 2023

May 17, 2023

Assessing the Impact of Context Inference Error and Partial Observability on RL Methods for Just-In-Time Adaptive Interventions
Karine Karine, Predrag Klasnja, Susan A. Murphy, Benjamin M. Marlin
Reinforcement Learning Global Impact Policy Gradient Partial Observability Adaptive Intervention Intervention Policy RL Method

May 12, 2023

S-REINFORCE: A Neuro-Symbolic Policy Gradient Approach for Interpretable Reinforcement Learning
Rajdeep Dutta, Qincheng Wang, Ankur Singh, Dhruv Kumarjiguda, Li Xiaoli, Senthilnath Jayavelu
Reinforcement Learning Policy Gradient Dynamic Decision Interpretable Policy Symbolic Policy

Policy Gradient

Papers

A Single-Loop Deep Actor-Critic Algorithm for Constrained Reinforcement Learning with Provable Convergence

Achieving Fairness in Multi-Agent Markov Decision Processes Using Reinforcement Learning

Efficient Diffusion Policies for Offline Reinforcement Learning

On the Linear Convergence of Policy Gradient under Hadamard Parameterization

Solving Robust MDPs through No-Regret Dynamics

Policy Gradient Algorithms for Robust MDPs with Non-Rectangular Uncertainty Sets

Policy Optimization for Continuous Reinforcement Learning

DoMo-AC: Doubly Multi-step Off-policy Actor-Critic Algorithm

A Policy Gradient Method for Confounded POMDPs

Decision-Aware Actor-Critic with Function Approximation and Theoretical Guarantees

Adaptive Policy Learning to Additional Tasks

Policy Learning based on Deep Koopman Representation

Proximal Policy Gradient Arborescence for Quality Diversity Reinforcement Learning

Training Diffusion Models with Reinforcement Learning

Off-Policy Average Reward Actor-Critic with Deterministic Policy Search

On First-Order Meta-Reinforcement Learning with Moreau Envelopes

Deep Metric Tensor Regularized Policy Gradient

Client Selection for Federated Policy Optimization with Environment Heterogeneity

Assessing the Impact of Context Inference Error and Partial Observability on RL Methods for Just-In-Time Adaptive Interventions

S-REINFORCE: A Neuro-Symbolic Policy Gradient Approach for Interpretable Reinforcement Learning