Reinforcement-learning-with-tensorflow icon indicating copy to clipboard operation
Reinforcement-learning-with-tensorflow copied to clipboard

Published 20 hours ago • verified publisher icon MorvanZhou

DQN的代码中，计算q_target时未考虑done为true的情况

Open ananasfl opened this issue 3 years ago • 1 comments

请问Morvan, DQN的代码中，计算q_target时，是否未考虑done为True的情况，即q_target = Reward? 存储在Replay memory中的经验也未包含done。请问为什么呢？

ananasfl avatar Apr 20 '22 06:04 ananasfl

请问有想到怎么考虑done=True的情况吗，如果在memory里存储经验包含done，那怎么解决随机取batch_size得到两个及以上done的情况？

ccconquer avatar May 31 '24 09:05 ccconquer