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
At the moment we're missing some basic training stuff like learning rate schedule. We are also missing some standard RL practices like observation clipping and reward clipping an option for terminal masking. They all should be fairly simple to implement and hopefully should give us small performance gains.
The text was updated successfully, but these errors were encountered:
riley-mld
changed the title
Basic training/architecture stuff (learning rate schedule, observation clipping, reward clipping, terminal masking option, ...)
Missing standard/basic training features
Mar 29, 2022
At the moment we're missing some basic training stuff like learning rate schedule. We are also missing some standard RL practices like observation clipping and reward clipping an option for terminal masking. They all should be fairly simple to implement and hopefully should give us small performance gains.
The text was updated successfully, but these errors were encountered: