search

marlbenchmark / on-policy

This is the official implementation of Multi-Agent PPO (MAPPO).
https://sites.google.com/view/mappo
MIT License
1.27k stars 292 forks source link

HAPPO为何与mappo的policy文件完全相同 #90

Closed strivebfq closed 11 months ago

strivebfq commented 1 year ago

作者你好，我在算法目录下看到HAPPO与mappo的policy文件完全相同，Happo一般为序贯决策，为何没有在Policy中体现，是否有额外的类或wrapper进行功能实现呢？