Stochastic Variance Reduction for Policy Gradient Estimation
read the original 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)
This paper has not been read by Pith yet.
Forward citations
Cited by 1 Pith paper
-
Input-Side Variance Suppression under Non-Normal Transient Amplification in Continuous-Control Reinforcement Learning
Non-normal transient amplification is an important contributor to closed-loop variance in RL, and input-side suppression can reduce downstream covariance without altering peak gain.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.