Reinforcement Learning on Humanoid
90,921,063Zero-Shot RewardOpen-Ended Neural Reward Functions
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Open-Ended Neural Reward FunctionsEvaluation Mode=Zero-shot2022.02 | 90,921,063 | — | |
| SMACbatch size=1000, seeds=52026.01 | 4,625 | — | |
| AC-SGDbatch size=1000, seeds=52026.01 | 4,539 | — | |
| AC-CGbatch size=1000, seeds=52026.01 | 3,175 | — | |
| AC-Adambatch size=1000, seeds=52026.01 | 3,105 | — | |
| MAMBA + Minimum AttentionSteps=30M2025.05 | 2,553 | — | |
| MAMBASteps=30M2025.05 | 2,453 | — | |
| CT-SACHyperparameter Configuration Rank=top2026.02 | 386.75 | — | |
| CT-SACHyperparameter Configuration Rank=second2026.02 | 379.75 | — | |
| CT-SACHyperparameter Configuration Rank=third2026.02 | 371.75 | — | |
| CT-TD3Hyperparameter Configuration Rank=top2026.02 | 337.23 | — | |
| CT-TD3Hyperparameter Configuration Rank=second2026.02 | 326.61 | — | |
| CT-TD3Hyperparameter Configuration Rank=third2026.02 | 295.94 | — | |
| AC-KFACbatch size=1000, seeds=52026.01 | 260.8 | — | |
| SACHyperparameter Configuration Rank=top2026.02 | 73.39 | — | |
| SACHyperparameter Configuration Rank=second2026.02 | 39.28 | — | |
| TD3Hyperparameter Configuration Rank=top2026.02 | 2.28 | — | |
| TD3Hyperparameter Configuration Rank=second2026.02 | 2.24 | — | |
| SACHyperparameter Configuration Rank=third2026.02 | 2.12 | — | |
| TD3Hyperparameter Configuration Rank=third2026.02 | 2.11 | — | |
| q-LearningHyperparameter Configuration Rank=top2026.02 | 1.81 | — | |
| q-LearningHyperparameter Configuration Rank=second2026.02 | 1.59 | — | |
| TRPOHyperparameter Configuration Rank=top2026.02 | 1.39 | — | |
| TRPOHyperparameter Configuration Rank=second2026.02 | 1.33 | — | |
| TRPOHyperparameter Configuration Rank=third2026.02 | 1.3 | — | |
| q-LearningHyperparameter Configuration Rank=third2026.02 | 1.28 | — | |
| PPOHyperparameter Configuration Rank=top2026.02 | 1.18 | — | |
| PPOHyperparameter Configuration Rank=second2026.02 | 1.17 | — | |
| CPPOHyperparameter Configuration Rank=top2026.02 | 1.16 | — | |
| PPOHyperparameter Configuration Rank=third2026.02 | 1.11 | — | |
| CPPOHyperparameter Configuration Rank=second2026.02 | 1.08 | — | |
| CPPOHyperparameter Configuration Rank=third2026.02 | 1.04 | — |