Robotics Control on DeepMind Control Suite Proprioceptive DMC-P
0.96TD3 Normalized ScoreULD
Evaluation results
| Method | Links | |
|---|---|---|
| ULDObservation Space=Vector, Action Space=Continuous, Training Steps=500k steps2026.02 | 0.96 | |
| TD-MPC2Observation Space=Vector, Action Space=Continuous, Training Steps=500k steps2026.02 | 0.95 | |
| TD7Observation Space=Vector, Action Space=Continuous, Training Steps=500k steps2026.02 | 0.87 | |
| DreamerV3Observation Space=Vector, Action Space=Continuous, Training Steps=500k steps2026.02 | 0.82 | |
| PPOObservation Space=Vector, Action Space=Continuous, Training Steps=500k steps2026.02 | 0.71 |