Skip to content
Discussion options

You must be logged in to vote

标准PPO也需要旧策略算 ratio。actor_old_model 就是每隔几步把当前 actor_model 的权重拷过去,专门用来提供 old_logp

kl_ref 当前策略和参考模型 ref_model 的 KL,别离最初的参考策略太远”的额外约束。

只是用 log 域计算更稳不会爆数,所以看起来是先 sum 再除,数学等价。

Replies: 6 comments 2 replies

Comment options

You must be logged in to vote
1 reply
@ciaoyizhen
Comment options

Answer selected by ciaoyizhen
Comment options

You must be logged in to vote
0 replies
Comment options

You must be logged in to vote
0 replies
Comment options

You must be logged in to vote
0 replies
Comment options

You must be logged in to vote
0 replies
Comment options

You must be logged in to vote
1 reply
@ciaoyizhen
Comment options

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Category
Q&A
Labels
None yet
2 participants