Offline Reinforcement Learning on D4RL Maze2D
155.8Return (UMaze)HD
Evaluation Results
| Method | Links | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| HD2025.10 | 155.8 | — | — | — | — | — | — | 135.6 | 128.4 | 139.9 | — | — | |
| DQL*2025.10 | 140.6 | — | — | — | — | — | — | 152 | 186.8 | 159.8 | — | — | |
| MDD2025.10 | 138.3 | — | — | — | — | — | — | 154.5 | 206.1 | 166.3 | — | — | |
| DV2025.10 | 136.6 | — | — | — | — | — | — | 150.7 | 203.6 | 163.6 | — | — | |
| SAC-N2026.03 | 130.59 | — | — | — | — | — | — | 88.61 | 204.76 | 141.32 | — | 423.96 | |
| Diffuser2025.10 | 113.9 | — | — | — | — | — | — | 121.5 | 123 | 119.5 | — | — | |
| ReBRAC2026.03 | 106.87 | — | — | — | — | — | — | 105.11 | 78.33 | 96.77 | — | 290.31 | |
| A2PRdataset_size=10k samples2026.01 | 100.6 | — | — | — | 125.5 | 108.7 | 155.9 | 156 | 97.3 | 111.4 | 35.7 | — | |
| EDAC2026.03 | 95.26 | — | — | — | — | — | — | 57.04 | 95.6 | 82.64 | — | 247.9 | |
| C4dataset_size=10k samples2026.01 | 87.8 | — | — | — | 130.3 | 145.9 | 166.1 | 122.3 | 152.9 | 126.9 | 83 | — | |
| DR3dataset_size=10k samples2026.01 | 76.9 | — | — | — | 117.6 | 85.7 | 133.5 | 93.2 | 82.2 | 89.4 | 37 | — | |
| TSRLdataset_size=10k samples2026.01 | 76.9 | — | — | — | 138 | 119.9 | 160 | 118 | 140 | 115.7 | 57 | — | |
| SfBC2025.10 | 73.9 | — | — | — | — | — | — | 73.8 | 74.4 | 74 | — | — | |
| SORLdataset_size=10k samples2026.01 | 72.5 | — | — | — | 63.4 | 119.4 | 115.4 | 96.4 | 58.3 | 83.3 | 58 | — | |
| IQLdataset_size=10k samples2026.01 | 68.6 | — | — | — | 103.6 | 100.1 | 181.7 | 104.4 | 103.2 | 103.6 | 63.6 | — | |
| AWAC2026.03 | 65.65 | — | — | — | — | — | — | 84.63 | 215.5 | 121.92 | — | 365.78 | |
| GEM2026.03 | 59.03 | — | — | — | — | — | — | 117.61 | 90.16 | 88.93 | — | 266.8 | |
| BPPOdataset_size=10k samples2026.01 | 58.9 | — | — | — | 51.4 | 81.2 | 217.2 | 132 | 228.1 | 120.6 | 75.4 | — | |
| IDQL2025.10 | 57.9 | — | — | — | — | — | — | 89.5 | 90.1 | 79.2 | — | — | |
| DOGEdataset_size=10k samples2026.01 | 52.8 | — | — | — | 115.1 | 100.9 | 149.3 | 127.7 | 126.5 | 106.1 | 70.1 | — | |
| Decision TransformerAugmentation=GTA, Number of evaluations=100, Number of seeds=42024.05 | 50.19 | — | — | — | — | — | — | 49.6 | 25.49 | 41.76 | — | — | |
| IQL2026.03 | 42.11 | — | — | — | — | — | — | 34.85 | 61.72 | 46.23 | — | 138.68 | |
| Decision TransformerAugmentation=None, Number of evaluations=100, Number of seeds=42024.05 | 39.43 | — | — | — | — | — | — | 55.24 | 18.49 | 37.72 | — | — | |
| Decision TransformerAugmentation=S4RL, Number of evaluations=100, Number of seeds=42024.05 | 39.38 | — | — | — | — | — | — | 55.05 | 18.69 | 37.71 | — | — | |
| CQLdataset_size=10k samples2026.01 | 35.9 | — | — | — | 74.1 | 38 | 78.5 | 80.1 | 42.7 | 57.4 | 52.3 | — | |
| TD3+BC2026.03 | 29.41 | — | — | — | — | — | — | 59.45 | 97.1 | 61.99 | — | 185.96 | |
| DT2026.03 | 18.08 | — | — | — | — | — | — | 31.71 | 35.66 | 28.48 | — | 85.45 | |
| LNdataset_size=10k samples2026.01 | 17.9 | — | — | — | 86.5 | 68.8 | 105.8 | 107.3 | 115.4 | 74.8 | 22.1 | — | |
| TD3BCdataset_size=10k samples2026.01 | 13.5 | — | — | — | 104.5 | 20.8 | 91.9 | 76.6 | 28.5 | 49.7 | 12.3 | — | |
| BC2025.10 | 3.8 | — | — | — | — | — | — | 30.3 | 5 | 13 | — | — | |
| CQL2026.03 | -8.9 | — | — | — | — | — | — | 86.11 | 23.75 | 33.65 | — | 100.96 | |
| BAILCategory=Model-Free2021.10 | — | — | — | — | — | 41.7 | — | — | — | — | — | — | |
| BCCategory=Baseline2021.10 | — | -3.2 | -0.5 | -1.7 | -6.9 | 2.7 | -0.3 | — | — | — | — | — | |
| BEARCategory=Model-Free2021.10 | — | 65.7 | — | 81 | — | — | 133 | — | — | — | — | — | |
| BRAC-PCategory=Model-Free2021.10 | — | — | — | — | 51.5 | — | — | — | — | — | — | — | |
| BRAC-vCategory=Model-Free2021.10 | — | — | 70.6 | — | — | — | — | — | — | — | — | — | |
| COMBOCategory=Model-Based2021.10 | — | 76.4 | 68.5 | 14.1 | — | 84.2 | — | — | — | — | — | — | |
| MOPOCategory=Model-Based2021.10 | — | — | — | — | — | — | 36.8 | — | — | — | — | — | |
| Repb-SDECategory=Model-Based2021.10 | — | — | — | — | 94.3 | — | — | — | — | — | — | — | |
| ROMI-BCQCategory=Proposed2021.10 | — | 139.5 | 82.4 | 83.1 | 98.3 | 102.6 | 124 | — | — | — | — | — |