hans / hans/rlcomp

Batch Norm issues

Open
#2 1 comment 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
18
Forks
2
PR merge metrics
No merged PRs in 30d

Description

Hello there!

I am trying to implement DDPG as well, but I am having a hard time trying to add batch norm to the actor and critic networks, i get really bad results with bn (diverging networks, or having trouble converging...). Did you managed to get a working version with bn (results at least as good as without?).

I am pretty new to neural networks, so I may have missed something...

(I am using tensorflow as-well)

Thanks.

Contributor guide

No contributing guide indexed for this repository

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.