Skip to content

lezhang-thu/bigger-better-faster-SAC

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

12 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Generalizing soft actor-critic algorithms to discrete action spaces

This repository implements the SAC-BBF agent in JAX (to be presented in PRCV 2024), building on BBF.

Results

  1. SAC-BBF improves the previous state-of-the-art interquartile mean (IQM) from 1.045 to 1.088.
  2. For ATARI 100K, SAC-BBF is also the only model-free algorithm with a super-human level using only replay ratio 2.

Setup

To install the repository, simply run pip install -r requirements.txt. Tested only on Python 3.10.

Training

To run a BBF agent locally, run

bash run-cuda0.sh

References

About

An external release of bigger-better-faster + SAC

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published