Find the optimal epsilon decay for decay of 1000 and 3000 epochs
Cross validation for epsilon decay with big decay. 20% 10% and 5% of the inital epsilon
| | L0 | L1 | L2 | L3 |
|---|---|---|---|---|
| learning rate | 0.2 | 0.2 | 0.2 | 0.2 |
| | E0 |
|---|---|
| epsilon | 0.05 |
| | ED0 | ED1 | ED2 |
|---|---|---|---|
| epsilon decay | decay-1000-20 | decay-1000-10 | decay-1000-05 |
| | ED3 | ED4 | ED5 |
| | decay-3000-20 | decay-3000-10 | decay-3000-05 |
| | D0 |
|---|---|
| discount | 0.3 |
| | M0 |
|---|---|
| mapping | non-linear-3 |
| | R0 |
|---|---|
| reward handler | can-see |
| | F0 |
|---|---|
| fetch mode | eager |