Sample Efficient Policy Gradient Methods with Recursive Variance Reduction

TL;DR

This paper introduces SRVR-PG, a new policy gradient method that significantly reduces sample complexity in reinforcement learning by employing recursive variance reduction, and demonstrates its effectiveness through experiments.

Contribution

The paper proposes SRVR-PG, a novel policy gradient algorithm with improved sample complexity, and a variant with parameter exploration, advancing reinforcement learning efficiency.

Findings

SRVR-PG achieves $O(1/\epsilon^{3/2})$ sample complexity.

The variant with parameter exploration enhances initial policy sampling.

Numerical experiments validate the improved performance on control problems.

Abstract

Improving the sample efficiency in reinforcement learning has been a long-standing research problem. In this work, we aim to reduce the sample complexity of existing policy gradient methods. We propose a novel policy gradient algorithm called SRVR-PG, which only requires $O(1/\epsilon^{3/2})$ episodes to find an $\epsilon$-approximate stationary point of the nonconcave performance function $J(\boldsymbol{\theta})$ (i.e., $\boldsymbol{\theta}$ such that $\|\nabla J(\boldsymbol{\theta})\|_2^2\leq\epsilon$). This sample complexity improves the existing result $O(1/\epsilon^{5/3})$ for stochastic variance reduced policy gradient algorithms by a factor of $O(1/\epsilon^{1/6})$. In addition, we also propose a variant of SRVR-PG with parameter exploration, which explores the initial policy parameter from a prior probability distribution. We conduct numerical experiments on classic control problems in reinforcement learning to validate the performance of our proposed algorithms.