Generalized Policy Improvement Algorithms with Theoretically Supported Sample Reuse

Code

• github.com/jqueeney/gpi
  OfficialIn papertf★ 2
• github.com/jqueeney/geppo
  tf★ 28

Abstract

We develop a new class of model-free deep reinforcement learning algorithms for data-driven, learning-based control. Our Generalized Policy Improvement algorithms combine the policy improvement guarantees of on-policy methods with the efficiency of sample reuse, addressing a trade-off between two important deployment requirements for real-world control: (i) practical performance guarantees and (ii) data efficiency. We demonstrate the benefits of this new class of algorithms through extensive experimental analysis on a broad range of simulated control tasks.

Tasks

Reproductions