You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
From the code, it is more like a DDPG algorithms but not PPO algorithms. Besides, for the auto-encoder part, this code is not use auto-encoder to extract features of input image, which looks use encoder as the CNNs. Why not learn and update the weight of these CNNs with actor and critics network? Why need to train these CNNs separately?
The text was updated successfully, but these errors were encountered:
From the code, it is more like a DDPG algorithms but not PPO algorithms. Besides, for the auto-encoder part, this code is not use auto-encoder to extract features of input image, which looks use encoder as the CNNs. Why not learn and update the weight of these CNNs with actor and critics network? Why need to train these CNNs separately?
The text was updated successfully, but these errors were encountered: