List your company

PointGoal2 on Safety-Gymnasium

67Normalized RewardBCQ-Lag
-2.378415.633333.64551.6567Apr 29, 2026

Evaluation results

MethodLinks
BCQ-Lag2026.04673.18
BEAR-Lag2026.04673.11
DT + SASBackbone=DT, Alignment strategy (SAS)=Lyapunov-Guided2026.04651.78
CDTBackbone=CDT2026.04591.34
BC-All2026.04541.97
CDT+SASAlignment=SAS, Base Model=CDT2026.045294
CDT + SASBackbone=CDT, Alignment strategy (SAS)=Lyapunov-Guided2026.04520.94
CPQ2026.04401.31
COptiDICE2026.04381.92
FISOR2026.043035
BC-Safe2026.04290.78
DCRL2026.04280.26
BCQ-Lagcost thresholds=3, evaluation episodes=20, random seeds=32026.040.673.18
BEAR-Lagcost thresholds=3, evaluation episodes=20, random seeds=32026.040.673.11
DT + SAScost thresholds=3, evaluation episodes=20, random seeds=32026.040.651.78
CDTcost thresholds=3, evaluation episodes=20, random seeds=32026.040.591.34
BC-Allcost thresholds=3, evaluation episodes=20, random seeds=32026.040.541.97
CDT + SAScost thresholds=3, evaluation episodes=20, random seeds=32026.040.520.94
CPQcost thresholds=3, evaluation episodes=20, random seeds=32026.040.41.31
COptiDICEcost thresholds=3, evaluation episodes=20, random seeds=32026.040.381.92
BC-Safecost thresholds=3, evaluation episodes=20, random seeds=32026.040.290.78