Reinforcement Learning Convergence on FrozenLake
27Median Episodes to ConvergeIn-Context
Evaluation Results
| Method | Links | |
|---|---|---|
| In-Contextseeds=122026.06 | 27 | |
| UCB-VIseeds=122026.06 | 46 | |
| Q-Learningseeds=122026.06 | 1,014 |
Loading the SOTA2 catalog…
| Method | Links | |
|---|---|---|
| In-Contextseeds=122026.06 | 27 | |
| UCB-VIseeds=122026.06 | 46 | |
| Q-Learningseeds=122026.06 | 1,014 |