Current plans or progress for TRPO and PPO

facebookresearch / ReAgent

A platform for Reasoning systems (Reinforcement Learning, Contextual Bandits, etc.)

https://reagent.ai

BSD 3-Clause "New" or "Revised" License

3.56k stars 516 forks source link

Open balloch opened 4 years ago

balloch commented 4 years ago

Im just curious if there is any effort to add these policy gradient methods ?

czxttkl commented 4 years ago

PPO is on our TODO list but has a low priority awaiting for someone to pick up.