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

December 2, 2021

Architecting and Visualizing Deep Reinforcement Learning Models
Alexander Neuwirth, Derek Riley
Deep Reinforcement Learning Policy Gradient Architecture Design Deep Reinforcement Learning Agent Human Player

November 30, 2021

Global Convergence Using Policy Gradient Methods for Model-free Markovian Jump Linear Quadratic Control
Santanu Rathod, Manoj Bhadu, Abir De
Reinforcement Learning Policy Gradient Global Convergence Natural Policy Gradient Markov Jump Gradient Based Policy

November 25, 2021

Distributed Policy Gradient with Variance Reduction in Multi-Agent Reinforcement Learning
Xiaoxiao Zhao, Jinlong Lei, Li Li, Jie Chen
Multi Agent Reinforcement Learning Policy Gradient Optimal Policy Stochastic Optimization Variance Reduction Multi Agent Policy Performance Function

November 23, 2021

Status-quo policy gradient in Multi-Agent Reinforcement Learning
Pinkesh Badjatiya, Mausoom Sarkar, Nikaash Puri, Jayakumar Subramanian, Abhishek Sinha, Siddharth Singh, Balaji Krishnamurthy
Reinforcement Learning Multi Agent Reinforcement Learning Policy Gradient Matrix Game Multi Agent Problem

November 22, 2021

Policy Gradient and Actor-Critic Learning in Continuous Time and Space: Theory and Algorithms
Yanwei Jia, Xun Yu Zhou
Reinforcement Learning Practical Algorithm Policy Gradient Theoretical Understanding Stochastic Approximation Actor Critic Algorithm Continuous Time Policy Parameterization

November 19, 2021

November 16, 2021

Off-Policy Actor-Critic with Emphatic Weightings
Eric Graves, Ehsan Imani, Raksha Kumaraswamy, Martha White
Policy Gradient Policy Actor Critic Linear Weighting Policy Policy Gradient

November 12, 2021

Guided Sampling-based Evolutionary Deep Neural Network for Intelligent Fault Diagnosis
Arun K. Sharma, Nishchal K. Verma
Deep Learning Model Policy Gradient DNN Architecture Evolutionary Approach Intelligent Fault Diagnosis Evolutionary Deep Learning

November 6, 2021

Time Discretization-Invariant Safe Action Repetition for Policy Gradient Methods
Seohong Park, Jaekyeom Kim, Gunhee Kim
Reinforcement Learning Policy Gradient Time Matter Action Repetition Delta Method

November 4, 2021

Global Optimality and Finite Sample Analysis of Softmax Off-Policy Actor Critic under State Distribution Mismatch
Shangtong Zhang, Remi Tachet, Romain Laroche
Policy Gradient Stochastic Approximation Global Optimality Distribution Matching Finite Sample Policy Actor Critic

November 3, 2021

Proximal Policy Optimization with Continuous Bounded Action Space via the Beta Distribution
Irving G. B. Petrazzini, Eric A. Antonelo
Reinforcement Learning Policy Gradient Action Space Proximal Policy Optimization Stochastic Policy Beta Distribution Gaussian Policy

Policy Gradient

Papers

Architecting and Visualizing Deep Reinforcement Learning Models

Global Convergence Using Policy Gradient Methods for Model-free Markovian Jump Linear Quadratic Control

Distributed Policy Gradient with Variance Reduction in Multi-Agent Reinforcement Learning

Status-quo policy gradient in Multi-Agent Reinforcement Learning

Policy Gradient and Actor-Critic Learning in Continuous Time and Space: Theory and Algorithms

Policy Gradient Approach to Compilation of Variational Quantum Circuits

Learn Quasi-stationary Distributions of Finite State Markov Chain

Off-Policy Actor-Critic with Emphatic Weightings

Guided Sampling-based Evolutionary Deep Neural Network for Intelligent Fault Diagnosis

Time Discretization-Invariant Safe Action Repetition for Policy Gradient Methods

Global Optimality and Finite Sample Analysis of Softmax Off-Policy Actor Critic under State Distribution Mismatch

Proximal Policy Optimization with Continuous Bounded Action Space via the Beta Distribution