List your company

Four Rooms on MiniGrid

88.7Average Pass RatePolychromic PPO
11.2231.33551.4571.565Sep 29, 2025

Evaluation results

MethodLinks
Polychromic PPOevaluation_protocol=pass@1, initial-state perturbations=large2025.0988.7
Poly-PPO w/ UCBevaluation_protocol=pass@1, initial-state perturbations=large2025.0987.2
REINFORCE w/ Baselineevaluation_protocol=pass@1, initial-state perturbations=large2025.0982.7
REINFORCE w/ UCBevaluation_protocol=pass@1, initial-state perturbations=large2025.0981.5
Pretrained policyevaluation_protocol=pass@1, initial-state perturbations=large2025.0965
PPOevaluation_protocol=pass@1, initial-state perturbations=large2025.0915.3
PPO w/ UCBevaluation_protocol=pass@1, initial-state perturbations=large2025.0914.2