Skip to content

Null-space preference critic: dual-critic PPO for uwlab_rl, plus grad-clip and wandb-resume fixes - #43

Draft
daphnechen wants to merge 7 commits into
UW-Lab:mainfrom
daphnechen:fix/pref-critic-clip
Draft

daphnechen wants to merge 7 commits into
UW-Lab:mainfrom
daphnechen:fix/pref-critic-clip

Resume the same wandb run across preemption restarts

64768df
Select commit
Loading
Failed to load commit list.