Deep-rl-mxnet icon indicating copy to clipboard operation
Deep-rl-mxnet copied to clipboard

Published 20 hours ago • verified publisher icon ZhengXinyue

DDPG/TD3 action saturation

Open wujingda opened this issue 2 years ago • 0 comments

Hi,

I found the DDPG/TD3 algorithms can easily lead to the action saturation (to maximum value) when training with tasks with more than one action. I noticed that you had ever experienced this problem and discussed it with others in 2019. Thus, I would like to kindly ask if you have addressed the problem? Thank you very much!

wujingda avatar May 27 '22 15:05 wujingda