rl_games icon indicating copy to clipboard operation
rl_games copied to clipboard

question about central value

Open JoseBarreiros-TRI opened this issue 1 year ago • 4 comments

@Denys88 thank you for the nice repo. I noticed you use a central value network when using asymmetric actor-critic. Could you please elaborate on what the central value net is exactly doing? Is this just the critic net?

JoseBarreiros-TRI avatar Mar 19 '24 20:03 JoseBarreiros-TRI

yes it is a critic network but with additional inputs. For example policy can have only a few parameters as obs but critic can have the whole world state because we don't use critic during inference.

Denys88 avatar Mar 21 '24 16:03 Denys88