Deep-reinforcement-learning-with-pytorch icon indicating copy to clipboard operation
Deep-reinforcement-learning-with-pytorch copied to clipboard

Published 20 hours ago • verified publisher icon sweetice

I dont think PPO pendulum is converging

Open Bigpig4396 opened this issue 5 years ago • 4 comments

Bigpig4396 avatar May 27 '19 17:05 Bigpig4396

Yes, the problem is that the activation function is chosen incorrectly.

KT27-A avatar Sep 28 '19 08:09 KT27-A

I don't think this repo implement the PPO correctly either

HuangHaoyu1997 avatar Mar 08 '20 17:03 HuangHaoyu1997

change the activation function relu to tanh

NanJuni avatar Mar 18 '20 13:03 NanJuni

right,change relu to tanh in actor network

wiluen avatar May 07 '21 08:05 wiluen