Reinforcement Learning on Halfcheetah v5
13,996.2Average ReturnSAC+TQC
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| SAC+TQCAlgorithm=SAC, Critic=TQC2026.02 | 13,996.2 | — | — | |
| TD3+DBC(*)Algorithm=TD3, Critic=DBC2026.02 | 13,787.8 | — | — | |
| QVPO+DBC(*)Algorithm=QVPO, Critic=DBC2026.02 | 13,182.1 | — | — | |
| SAC+DBC(*)Algorithm=SAC, Critic=DBC2026.02 | 13,120.2 | — | — | |
| SAC+DSACAlgorithm=SAC, Critic=DSAC2026.02 | 12,974 | — | — | |
| SAC+CDQAlgorithm=SAC, Critic=CDQ2026.02 | 12,776.4 | — | — | |
| Full State SACState Dim Removed=None2026.02 | 12,400 | — | — | |
| Full State SACState Dim Removed=None2026.02 | 12,400 | — | — | |
| QVPO+CDQAlgorithm=QVPO, Critic=CDQ2026.02 | 12,103.9 | — | — | |
| TD3+CDQAlgorithm=TD3, Critic=CDQ2026.02 | 11,679 | — | — | |
| SAC+IQNAlgorithm=SAC, Critic=IQN2026.02 | 10,930.2 | — | — | |
| Recurrent SACState Dim Removed=32026.02 | 10,057 | — | — | |
| Recurrent SACState Dim Removed=32026.02 | 9,875 | — | — | |
| TFM-S3-TD3Algorithm=TFM-S3-TD3, Variant=proposed2026.04 | 9,863.4 | 40.5 | 62.7 | |
| CQLState Dim Removed=32026.02 | 9,748 | — | — | |
| CQLState Dim Removed=32026.02 | 9,748 | — | — | |
| Random Search TD3Algorithm=Random Search, Candidates=322026.04 | 9,686.3 | 38.4 | 66 | |
| TD3Algorithm=TD3, Variant=vanilla2026.04 | 9,624.1 | 39.6 | 71.3 | |
| TFM-S3-TD3Algorithm=TFM-S3-TD3, Variant=One shot2026.04 | 9,576.9 | 37.7 | 68 | |
| T-REXState Dim Removed=32026.02 | 9,120 | — | — | |
| TrexState Dim Removed=32026.02 | 9,106 | — | — | |
| Causal PBRSState Dim Removed=32026.02 | 8,954 | — | — | |
| Causal PBRSState Dim Removed=32026.02 | 8,949 | — | — | |
| BaselineState Dim Removed=32026.02 | 8,931 | — | — | |
| SAC BaselineState Dim Removed=32026.02 | 8,931 | — | — | |
| SAC+VFAlgorithm=SAC, Critic=VF2026.02 | 8,199.5 | — | — | |
| Recurrent SACState Dim Removed=12026.02 | 7,957 | — | — | |
| Recurrent SACState Dim Removed=82026.02 | 7,939 | — | — | |
| Recurrent SACState Dim Removed=82026.02 | 7,939 | — | — | |
| Recurrent SACState Dim Removed=12026.02 | 7,378 | — | — | |
| SAC+VDAlgorithm=SAC, Critic=VD2026.02 | 4,284.8 | — | — | |
| T-REXState Dim Removed=12026.02 | 2,059 | — | — | |
| TrexState Dim Removed=12026.02 | 2,057 | — | — | |
| Causal PBRSState Dim Removed=12026.02 | 2,055 | — | — | |
| Causal PBRSState Dim Removed=12026.02 | 2,051 | — | — | |
| CQLState Dim Removed=12026.02 | 2,039 | — | — | |
| CQLState Dim Removed=12026.02 | 2,039 | — | — | |
| BaselineState Dim Removed=12026.02 | 2,013 | — | — | |
| SAC BaselineState Dim Removed=12026.02 | 2,013 | — | — | |
| Causal PBRSState Dim Removed=82026.02 | 1,103 | — | — | |
| Causal PBRSState Dim Removed=82026.02 | 1,103 | — | — | |
| SAC BaselineState Dim Removed=82026.02 | 896 | — | — | |
| BaselineState Dim Removed=82026.02 | 895 | — | — | |
| T-REXState Dim Removed=82026.02 | 615 | — | — | |
| TrexState Dim Removed=82026.02 | 612 | — | — | |
| CQLState Dim Removed=82026.02 | 373 | — | — | |
| CQLState Dim Removed=82026.02 | 373 | — | — |