nikitasrivatsan / DeepLearningVideoGames

1.08k stars 215 forks source link

Asynchronous Methods for Deep Reinforcement Learning #9

Open Zeta36 opened 8 years ago

Zeta36 commented 8 years ago

Using a paper from Google DeepMind I've developed a new version of the DQN using threads exploration instead of memory replay as explain in here: http://arxiv.org/pdf/1602.01783v1.pdf

I used the one-step-Q-learingn pseudocode, and now we can train the Pong game in less than 20 hours and without any GPU or network distribution.