Accelerating Policy Gradient by Estimating Value Function from Prior Computation in Deep Reinforcement Learning

TL;DR

This paper proposes a method to improve sample efficiency in policy gradient reinforcement learning by leveraging prior computed value functions, combining them with learned estimates to reduce variance and accelerate training.

Contribution

It introduces a novel approach to incorporate prior value estimates into policy gradient methods, enhancing sample efficiency and performance in reinforcement learning tasks.

Findings

Prior value estimates improve sample efficiency.

Combining prior and learned value functions reduces gradient variance.

Method outperforms baseline in multiple RL environments.

Abstract

This paper investigates the use of prior computation to estimate the value function to improve sample efficiency in on-policy policy gradient methods in reinforcement learning. Our approach is to estimate the value function from prior computations, such as from the Q-network learned in DQN or the value function trained for different but related environments. In particular, we learn a new value function for the target task while combining it with a value estimate from the prior computation. Finally, the resulting value function is used as a baseline in the policy gradient method. This use of a baseline has the theoretical property of reducing variance in gradient computation and thus improving sample efficiency. The experiments show the successful use of prior value estimates in various settings and improved sample efficiency in several tasks.