Medium
7 / 527h 24m · 30 replays ↗
Can an agent learn to win across repeated games?
| Experiment | Reasoning | Sessions | Games / session | Final score | |
|---|---|---|---|---|---|
| v6 | Medium | 1 | 30 | 7 / 52 | Replays ↗ |
| v6 | Max | 1 | 30 | 13 / 52 | Replays ↗ |
| v5 | Max | 1 | 8 + 2 | 6 / 52 | Replays ↗ |
| v3 | Medium | 8 | 8 + 2 | 5.9 / 52* | Replays ↗ |
30 games: best of all. 8+2: best of the last two. *Mean across completed sessions.
7h 24m · 30 replays ↗
12h · 30 replays ↗
Same 30 dungeons. Separate sessions. Twelve-hour allowance each.