DDPG, Tensorflow1, Tensorflow2
使用DDPG跑pendulam倒立摆
Tensorflow2网络模型：
Actor:
Critic:
TensorFlow1
actor 添加批量正则化，实验表明，更具有数值稳定性
Critic 多添加两个Dense层，把state调整到300，action调整到300，然后对应元素相加，而不是tensorflow2版本的concate.
Reference
-
Notifications
You must be signed in to change notification settings - Fork 1
Jian-Yin-Shine/DDPG_Tensorflow
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
About
DDPG, Tensorflow1, Tensorflow2
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published