Hindsight Experience Replay with Kronecker Product Approximate Curvature

TL;DR

This paper introduces a novel reinforcement learning algorithm combining Hindsight Experience Replay with Kronecker-Factored Approximate Curvature and TD3 to improve sample efficiency, convergence speed, and success rates in sparse reward environments.

Contribution

It proposes integrating natural gradient methods and Kronecker-factored curvature into HER with TD3, enhancing performance and training efficiency in reinforcement learning tasks.

Findings

Improved sample efficiency over standard HER.

Faster convergence in Mujoco environments.

Higher success rates in sparse reward tasks.

Abstract

Hindsight Experience Replay (HER) is one of the efficient algorithm to solve Reinforcement Learning tasks related to sparse rewarded environments.But due to its reduced sample efficiency and slower convergence HER fails to perform effectively. Natural gradients solves these challenges by converging the model parameters better. It avoids taking bad actions that collapse the training performance. However updating parameters in neural networks requires expensive computation and thus increase in training time. Our proposed method solves the above mentioned challenges with better sample efficiency and faster convergence with increased success rate. A common failure mode for DDPG is that the learned Q-function begins to dramatically overestimate Q-values, which then leads to the policy breaking, because it exploits the errors in the Q-function. We solve this issue by including Twin Delayed Deep Deterministic Policy Gradients(TD3) in HER. TD3 learns two Q-functions instead of one and it adds noise tothe target action, to make it harder for the policy to exploit Q-function errors. The experiments are done with the help of OpenAis Mujoco environments. Results on these environments show that our algorithm (TDHER+KFAC) performs better inmost of the scenarios