deep-reinforcement-learning_DDQN_PPO_HER icon indicating copy to clipboard operation
deep-reinforcement-learning_DDQN_PPO_HER copied to clipboard

Published 20 hours ago • verified publisher icon monoelh

Dueling networks in mlp_framework.py (duel-double-DQN,van Hasselt)

Open monoelh opened this issue 6 years ago • 0 comments

see dueling_mlp class in mlp_framework.py

I've problems with calculating the correct loss for both streams (Advantage and Value).

monoelh avatar Jan 15 '18 14:01 monoelh