Reinforcement Learning on frozen Small
0.99TimeMungojerrie
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Mungojerriestates=16, prod.=64, c=-1, ε=0.1, α=0.1, η=0.1, train-steps=500k2025.05 | 0.99 | — | — | |
| Q-learning with reduction (Bozkurt et al. 2020)states=16, prod.=64, c=-1, ε=0.1, α=0.1, η=0.1, train-steps=500k2025.05 | — | — | 9.88 | |
| Q-learning with reduction (Hahn et al. 2019)states=16, prod.=64, c=-1, ε=0.1, α=0.1, η=0.1, train-steps=500k2025.05 | — | 20.23 | — |