Matrix Game Strategy Learning on Matching Pennies
0Last Iterate ExploitabilityR-NaD
Evaluation Results
| Method | Links | |
|---|---|---|
| R-NaDseeds=10, steps=5000, annealing=none2026.06 | 0 | |
| MMDseeds=10, steps=5000, annealing=none2026.06 | 0 | |
| Optimistic MDseeds=10, steps=5000, annealing=none2026.06 | 0 | |
| GARIPseeds=10, steps=5000, annealing=none2026.06 | 0.0001 | |
| Fictitious playseeds=10, steps=5000, annealing=none2026.06 | 0.02 | |
| Self-play GAseeds=10, steps=5000, annealing=none2026.06 | 1.998 |