Bandit Regret Minimization on Checkpoint-based
427.7Mean DifferenceMF-UCB
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| MF-UCBFinal budget (Λ)=256000, Number of common seeds=1202026.05 | 427.7 | 1,125.4 | — | |
| DNCFinal budget (Λ)=256000, Number of common seeds=1202026.05 | 737.5 | 1,606.2 | — | |
| Low-ImprovingFinal budget (Λ)=256000, Number of common seeds=1202026.05 | 1,989.8 | 1,521.2 | — | |
| UCBFinal budget (Λ)=256000, Number of common seeds=1202026.05 | 2,356.3 | 2,831 | — | |
| Weak-FixedFinal budget (Λ)=256000, Number of common seeds=1202026.05 | 2,803.1 | 3,273.9 | — |