Reinforcement Learning on Maze Navigation approximate transition/Q-values (test)
2,016.5Average Performance ValuePrimitives+Macro
Evaluation results
| Method | Links | |
|---|---|---|
| Primitives+MacroAction set=Primitives + Macros, Implementation=Tabular2017.11 | 2,016.5 | |
| PrimitivesAction set=Primitives, Implementation=Tabular2017.11 | 2,355.44 | |
| Option-CriticImplementation=Tabular2017.11 | 2,788.52 | |
| Eigen-OptionsImplementation=Tabular2017.11 | 3,444.06 |