Stochastic Variance Reduction for Policy Gradient Estimation

Abstract

Recent advances in policy gradient methods and deep learning have demonstrated their applicability for complex reinforcement learning problems. However, the variance of the performance gradient estimates obtained from the simulation is often excessive, leading to poor sample efficiency. In this paper, we apply the stochastic variance reduced gradient descent (SVRG) to model-free policy gradient to significantly improve the sample-efficiency. The SVRG estimation is incorporated into a trust-region Newton conjugate gradient framework for the policy optimization. On several Mujoco tasks, our method achieves significantly better performance compared to the state-of-the-art model-free policy gradient methods in robotic continuous control such as trust region policy optimization (TRPO)

7 pages, 3 figures

References in corpus (4)

• Trust Region Policy Optimization
• Benchmarking Deep Reinforcement Learning for Continuous Control
• Stochastic Variance Reduction Methods for Policy Evaluation
• Improved Optimization of Finite Sums with Minibatch Stochastic Variance Reduced Proximal Iterations

Cited by in corpus (5)

• Sample Efficient Policy Gradient Methods with Recursive Variance Reduction
• An Improved Convergence Analysis of Stochastic Variance-Reduced Policy Gradient
• Stochastic Recursive Momentum for Policy Gradient Methods
• Variance Reduction for Deep Q-Learning using Stochastic Recursive Gradient
• Fault-Tolerant Federated Reinforcement Learning with Theoretical Guarantee