Offline Reinforcement Learning on D4RL Hopper v2 (test)
776.7Mean Episode RewardRGMDT
Evaluation Results
| Method | Links | |
|---|---|---|
| RGMDTSample Size=80,000 samples, Node Count=64 nodes2024.10 | 776.7 | |
| RGMDTSample Size=80,000 samples, Node Count=40 nodes2024.10 | 567.27 | |
| RFSample Size=800,000 samples, Node Count=64 nodes2024.10 | 489.92 | |
| RGMDTSample Size=800,000 samples, Node Count=40 nodes2024.10 | 467.27 | |
| CARTSample Size=800,000 samples, Node Count=64 nodes2024.10 | 460.9 | |
| RGMDTSample Size=800,000 samples, Node Count=64 nodes2024.10 | 458.7 | |
| CARTSample Size=800,000 samples, Node Count=40 nodes2024.10 | 458.47 | |
| RFSample Size=800,000 samples, Node Count=40 nodes2024.10 | 456.41 | |
| RFSample Size=80,000 samples, Node Count=64 nodes2024.10 | 452.89 | |
| ETSample Size=800,000 samples, Node Count=64 nodes2024.10 | 451.8 | |
| ETSample Size=80,000 samples, Node Count=64 nodes2024.10 | 448.98 | |
| CARTSample Size=80,000 samples, Node Count=64 nodes2024.10 | 448.85 | |
| CARTSample Size=80,000 samples, Node Count=40 nodes2024.10 | 443.47 | |
| ETSample Size=800,000 samples, Node Count=40 nodes2024.10 | 441.01 | |
| RFSample Size=80,000 samples, Node Count=40 nodes2024.10 | 345.81 | |
| ETSample Size=80,000 samples, Node Count=40 nodes2024.10 | 196.55 |