Reinforcement Learning Surrogate Modeling on HalfCheetah (test)
85Reward RatioCubist
Evaluation results
| Method | Links | ||
|---|---|---|---|
| CubistMNR=70, Model Size=7500±972026.07 | 85 | 0.042 | |
| ORCAIDMD=6, Model Size=3215±1302026.07 | 78 | 0.051 | |
| Decision TreeMD=20, Model Size=6032±1942026.07 | 47 | 0.072 | |
| RuleFitMNR=32, Model Size=721±212026.07 | 46 | 0.081 |