search

NVlabs / GA3C

Hybrid CPU/GPU implementation of the A3C algorithm for deep reinforcement learning.
BSD 3-Clause "New" or "Revised" License
652 stars 195 forks source link

Need an action trigger for 'press to continue' kind of situations #13

Open yumere opened 7 years ago

yumere commented 7 years ago

For some Atari games including Breakout, the environment sometimes waits for user input to continue (when the user looses a life, for example). Game play may be stuck forever if 'no-op' action is set in such situations. To prevent this, ProcessAgent may need an action sequence for repeated 'no-op' actions and should take an 'real' action if the queue length goes beyond the limit.

mbz commented 7 years ago

This is a good feature to add but in our experiments it looks unnecessary. Usually the agent figures out the importance of pressing the fire button quickly enough and at the beginning it usually gets pressed by random (at least in case of Breakout).