Continuous Control on MuJoCo Ant
6,336Average RewardTOP-TD3
Evaluation results
| Method | Links | |
|---|---|---|
| TOP-TD3Training steps=1M, Number of trials=102021.02 | 6,336 | |
| ND TOP-TD3Training steps=1M, Number of trials=102021.02 | 6,096 | |
| QR-TD3Training steps=1M, Number of trials=102021.02 | 5,642 | |
| CrossQ+FEMAintegration=FEMA2026.03 | 5,082.74 | |
| CrossQ2026.03 | 4,875.93 | |
| OACTraining steps=1M, Number of trials=102021.02 | 4,761 | |
| SAC+FEMAintegration=FEMA2026.03 | 4,757.64 | |
| ExpertNumber of expert trajectories=102021.06 | 4,700 | |
| IQ-LearnNumber of expert trajectories=102021.06 | 4,683 | |
| ValueDICENumber of expert trajectories=102021.06 | 4,507 | |
| TD3Training steps=1M, Number of trials=102021.02 | 4,242 | |
| SAC2026.03 | 4,174.66 | |
| GAILNumber of expert trajectories=102021.06 | 3,997 | |
| SACTraining steps=1M, Number of trials=102021.02 | 3,421 | |
| QAvatarTraining steps=500k steps2026.03 | 2,858.8 | |
| BCNumber of expert trajectories=102021.06 | 2,272 | |
| FTTraining steps=500k steps2026.03 | 1,644.3 | |
| CAT-SACTraining steps=500k steps2026.03 | 1,628.9 | |
| SACTraining steps=500k steps2026.03 | 1,620 | |
| EMAC2026.03 | 832.98 | |
| CMDTraining steps=500k steps2026.03 | 777.5 | |
| PARTraining steps=500k steps2026.03 | 737.6 | |
| PPO+FEMAintegration=FEMA2026.03 | 427.7 | |
| PPO2026.03 | 228.52 | |
| CATTraining steps=500k steps2026.03 | 17.1 | |
| RandomNumber of expert trajectories=102021.06 | -70 |