search

google / dopamine

Dopamine is a research framework for fast prototyping of reinforcement learning algorithms.
https://github.com/google/dopamine
Apache License 2.0
10.44k stars 1.37k forks source link

Modifying dopamine to accept continuous action spaces #53

Open crobarcro opened 5 years ago

crobarcro commented 5 years ago

Could you possibly comment on the difficult of getting dopamine to work with continuous action spaces? Is this something that could be done with a bit of effort, or are the agents completely incompatible with this type of action space? I wanted to have a go getting dopamine to work with the OpenAI Gym BipedalWalker environment. I managed to get all of the initialization and setup working, then realized an assumption of discreet action spaces was hard-coded into dopamine.

mgbellemare commented 5 years ago

It probably isn't easy, the agents are incompatible with these action spaces. This was probably the toughest choice for the first version – to support continuous actions or not. It would have significantly increased code complexity, so we decided to wait until we had a better sense of where to take Dopamine next.

crobarcro commented 5 years ago

Ok, thanks for this, I appreciate you taking the time to answer my question.