Deep-Reinforcement-Learning-Practice

Deep-Reinforcement-Learning-Practice copied to clipboard

Is it ok to run the model on gpu with a3c algorithm?. If you know a2c algorithm, please tell me is that better than a3c on gpu?. Thank you

nguyenviettuan96

Thanks for your contribution.

shtse8

Hello，I find in double dqn file, there is no updated parameters function like this： $$ \theta^{-}=\alpha \theta^{-}+(1-\alpha) \theta $$ can you tell me why and what differences between this two?...

ynuwm

请问对于连续控制任务，如果可选的动作action有多个（假设6个），PPO采用随机策略其actor最后一层的输出是什么？

davinca