Skip to content

Personal project to reproduce the deep deterministic policy gradient method

Notifications You must be signed in to change notification settings

bencottier/ddpg-replica

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

98 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Reproducing DDPG in Python with TensorFlow

A personal project to reproduce the deep deterministic policy gradient (DDPG) algorithm for continuous control under reinforcement learning (RL) [1]. It is considered reproduced if it obtains similar results to [1], or else achieves respectable average return in multiple popular continuous control toy environments.

The implementation is in Python using the TensorFlow library. It uses the OpenAI Gym API to handle environments, and is heavily based on the author's memory of the OpenAI Spinning Up library. However, as a rule no DDPG or related algorithm implementations are directly read while the project is in progress.

[1] Lillicrap, Timothy P., Jonathan J. Hunt, Alexander Pritzel, Nicolas Heess, Tom Erez, Yuval Tassa, David Silver, and Daan Wierstra. "Continuous control with deep reinforcement learning." arXiv preprint arXiv:1509.02971 (2015).

About

Personal project to reproduce the deep deterministic policy gradient method

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published