Giters

rlcode / reinforcement-learning

Minimal and Clean Reinforcement Learning Examples

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Batch size in A2C Cartpole

akileshbadrinaaraayanan opened this issue · comments

commented

Are there no variants of A2C with mini-batch update instead of training every time step? If yes, could you tell the pros and cons of such an approach?

Thanks,
Akilesh

commented

If you want, you can change it from one-step bootstrap to multi-step bootstrap. As I know, one-step bootstrap has lower variance and higher bias than multi-step bootstrap.