Reinforcement Learning Surrogate Modeling on Hopper H (test)
89Reward RatioORCAID
Evaluation results
| Method | Links | ||
|---|---|---|---|
| ORCAIDMD=4, Model Size=720±102026.07 | 89 | 0.051 | |
| CubistMNR=20, Model Size=850±62026.07 | 86 | 0.046 | |
| RuleFitMNR=28, Model Size=287±72026.07 | 85 | 0.077 | |
| Decision TreeMD=18, Model Size=4238±672026.07 | 17 | 0.056 |