https://github.com/godka/dcppg

Dual-clip Phasic Policy Gradient
https://github.com/godka/dcppg

deep-reinforcement-learning

Last synced: about 10 hours ago
JSON representation

Dual-clip Phasic Policy Gradient

• Host: GitHub
• URL: https://github.com/godka/dcppg
• Owner: godka
• Created: 2023-08-13T10:43:45.000Z (over 1 year ago)
• Default Branch: master
• Last Pushed: 2024-01-16T02:19:37.000Z (10 months ago)
• Last Synced: 2024-01-16T07:27:58.529Z (10 months ago)
• Topics: deep-reinforcement-learning
• Language: Python
• Homepage:
• Size: 6.84 KB
• Stars: 0
• Watchers: 2
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# DCPPG

Dual-clip Phasic Policy Gradient: the simplest implementation of DCPPG (dual-clip phasic policy gradient), which can be viewed as the fusion of Dual-clip PPO + PPG. It significantly improves the converge speed compared with DCPPO.

## Dual-clip PPO: 

Ye, Deheng, et al. "Mastering complex control in moba games with deep reinforcement learning." Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 34. No. 04. 2020.

## Phasic Policy Gradient: 

Cobbe, Karl W., et al. "Phasic policy gradient." International Conference on Machine Learning. PMLR, 2021.