List your company

Offline Reinforcement Learning on D4RL Summary v0 (Aggregate)

71.12Average Return (All Settings)ROAD
46.763253.086659.4165.7334Jun 2, 2021Mar 30, 2022Jan 25, 2023Nov 22, 2023Sep 18, 2024Jul 16, 2025May 14, 2026

Evaluation results

MethodLinks
ROADBackbone=PEX2026.0571.12—
Decision TransformerScore Type=Average Normalized Score2021.0669.274.7
Fixed Mixing RatioMixing Ratio=0.1, Backbone=PEX2026.0560.97—
Decreasing Mixing StrategyBackbone=PEX2026.0558.36—
BRBackbone=PEX2026.0557.47—
Fixed Mixing RatioMixing Ratio=0.2, Backbone=PEX2026.0556.91—
Fixed Mixing RatioMixing Ratio=0.4, Backbone=PEX2026.0556.53—
Fixed Mixing RatioMixing Ratio=0.0, Backbone=PEX2026.0556.09—
Fixed Mixing RatioMixing Ratio=0.3, Backbone=PEX2026.0555.32—
Uniform Mixing StrategyBackbone=PEX2026.0555.12—
Fixed Mixing RatioMixing Ratio=0.5, Backbone=PEX2026.0554.3—
CQLScore Type=Average Normalized Score2021.0654.263.9
BCScore Type=Average Normalized Score2021.0647.746.4
AWRScore Type=Average Normalized Score2021.06—31.3
BEARScore Type=Average Normalized Score2021.06—48.2
BRAC-vScore Type=Average Normalized Score2021.06—36.9