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
Our implementation of RelEnt currently works with trajectories of varying length. (This is because we rely on our collect_trajs util, which returns when an episode ends.)
By contrast, the RelEnt paper does all calculations under the assumption of a fixed trajectory length.
I'm not sure if this is problematic, but open this issue lest we forget to look into this.
The text was updated successfully, but these errors were encountered:
Our implementation of RelEnt currently works with trajectories of varying length. (This is because we rely on our collect_trajs util, which returns when an episode ends.)
By contrast, the RelEnt paper does all calculations under the assumption of a fixed trajectory length.
I'm not sure if this is problematic, but open this issue lest we forget to look into this.
The text was updated successfully, but these errors were encountered: