You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Since the implementation for divergence-free reward distance calculations will also live in this repo (as well as possible other reward function distance implementations), the repo should be renamed to reflect that. setup.py etc. will also need to be modified.
The text was updated successfully, but these errors were encountered:
Yeah, EPIC is definitely too restrictive. We were having a discussion on Slack for what to call these class of distances with @ejnnr and @JoarSkalse. I think we're currently leaning towards STARTC (STAndardised Reward Comparison) which is also available on PyPi. Full options:
Since the implementation for divergence-free reward distance calculations will also live in this repo (as well as possible other reward function distance implementations), the repo should be renamed to reflect that.
setup.py
etc. will also need to be modified.The text was updated successfully, but these errors were encountered: