List your company

Reinforcement Learning on MiniGrid BlockedUnlockPickup

80Success Rate (SR)Hand-Crafted
-3.218.44061.6May 27, 2026

Evaluation results

MethodLinks
Hand-CraftedTraining Episodes=3,0002026.0580
LLM One-ShotTraining Episodes=3,0002026.0570
LLM IterativeTraining Episodes=3,0002026.0560
No ShapingTraining Episodes=3,0002026.050
RNDTraining Episodes=3,0002026.050