Lazy fetch cross validation. 20k epochs.
Cross validation for lazy fetching
| | L0 | L1 | L2 |
|---|---|---|---|
| learning rate | 0.2 | 0.2 | 0.2 |
| | E0 | E1 | E2 | E3 |
|---|---|---|---|---|
| epsilon | 0.01 | 0.01 | 0.01 | 0.01 |
| | D0 | D1 |
|---|---|---|
| discount | 0.3 | 0.3 |
| | M0 |
|---|---|
| mapping | non-linear-3 |
| | R0 |
|---|---|
| reward handler | can-see |
| | F0 | F1 | F2 |
|---|---|---|---|
| fetch mode | lazy-s | lazy-m | lazy-l |