Offline Reinforcement Learning on Maze2D medium
179.2Normalized ReturnOURS
Evaluation Results
| Method | Links | |
|---|---|---|
| OURS2026.02 | 179.2 | |
| GPC-SAC2026.02 | 103.7 | |
| Real DatasetOffline RL Algorithm=IQL, Privacy Budget (epsilon)=no-DP2025.12 | 93.1 | |
| PrivORL-nOffline RL Algorithm=IQL, Privacy Budget (epsilon)=102025.12 | 90.7 | |
| PrivORL-nOffline RL Algorithm=IQL, Privacy Budget (epsilon)=12025.12 | 73.5 | |
| Real DatasetOffline RL Algorithm=EDAC, Privacy Budget (epsilon)=no-DP2025.12 | 73 | |
| PBRL2026.02 | 71.1 | |
| Real DatasetOffline RL Algorithm=TD3PLUSBC, Privacy Budget (epsilon)=no-DP2025.12 | 53.3 | |
| PrivORL-nOffline RL Algorithm=TD3PLUSBC, Privacy Budget (epsilon)=102025.12 | 50.4 | |
| PGMOffline RL Algorithm=IQL, Privacy Budget (epsilon)=102025.12 | 46.8 | |
| PrivORL-nOffline RL Algorithm=EDAC, Privacy Budget (epsilon)=102025.12 | 45.6 | |
| PGMOffline RL Algorithm=IQL, Privacy Budget (epsilon)=12025.12 | 35.5 | |
| PrivORL-nOffline RL Algorithm=TD3PLUSBC, Privacy Budget (epsilon)=12025.12 | 33.8 | |
| PrivORL-nOffline RL Algorithm=EDAC, Privacy Budget (epsilon)=12025.12 | 32 | |
| IQL2026.02 | 32 | |
| PrivSynOffline RL Algorithm=TD3PLUSBC, Privacy Budget (epsilon)=102025.12 | 31.6 | |
| PrivSynOffline RL Algorithm=TD3PLUSBC, Privacy Budget (epsilon)=12025.12 | 30 | |
| PATE-GANOffline RL Algorithm=IQL, Privacy Budget (epsilon)=102025.12 | 25.7 | |
| CQL2026.02 | 24.8 | |
| PGMOffline RL Algorithm=EDAC, Privacy Budget (epsilon)=102025.12 | 20.3 | |
| PATE-GANOffline RL Algorithm=TD3PLUSBC, Privacy Budget (epsilon)=102025.12 | 20 | |
| PATE-GANOffline RL Algorithm=IQL, Privacy Budget (epsilon)=12025.12 | 17.7 | |
| PATE-GANOffline RL Algorithm=EDAC, Privacy Budget (epsilon)=102025.12 | 15.5 | |
| PATE-GANOffline RL Algorithm=TD3PLUSBC, Privacy Budget (epsilon)=12025.12 | 15.3 | |
| PrePATE-GANOffline RL Algorithm=EDAC, Privacy Budget (epsilon)=102025.12 | 14.3 | |
| PGMOffline RL Algorithm=EDAC, Privacy Budget (epsilon)=12025.12 | 12.1 | |
| PATE-GANOffline RL Algorithm=EDAC, Privacy Budget (epsilon)=12025.12 | 10 | |
| PGMOffline RL Algorithm=TD3PLUSBC, Privacy Budget (epsilon)=102025.12 | 7.7 | |
| PGMOffline RL Algorithm=TD3PLUSBC, Privacy Budget (epsilon)=12025.12 | 4.3 | |
| PrivSynOffline RL Algorithm=IQL, Privacy Budget (epsilon)=102025.12 | 4 | |
| PrePATE-GANOffline RL Algorithm=EDAC, Privacy Budget (epsilon)=12025.12 | 2.3 | |
| PrivSynOffline RL Algorithm=IQL, Privacy Budget (epsilon)=12025.12 | 2 | |
| PrivSynOffline RL Algorithm=EDAC, Privacy Budget (epsilon)=12025.12 | 0.2 | |
| PrePATE-GANOffline RL Algorithm=TD3PLUSBC, Privacy Budget (epsilon)=12025.12 | -2.2 | |
| PrePATE-GANOffline RL Algorithm=IQL, Privacy Budget (epsilon)=12025.12 | -4.2 | |
| PrePATE-GANOffline RL Algorithm=TD3PLUSBC, Privacy Budget (epsilon)=102025.12 | -5.4 | |
| PrivSynOffline RL Algorithm=EDAC, Privacy Budget (epsilon)=102025.12 | -6.3 | |
| PrePATE-GANOffline RL Algorithm=IQL, Privacy Budget (epsilon)=102025.12 | -7 |