Bandit Regret Minimization on Wheel Bandit delta=0.9
7,654.6Final Cumulative Regret (Mean)Gradient-Laplace
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Gradient-Laplacek=5002026.05 | 7,654.6 | 430 | |
| Greedy-Laplacek=5002026.05 | 7,732.3 | 383.1 | |
| Gradient-Laplacek=1002026.05 | 20,747.6 | 12,895.4 | |
| NeuralLineark=N/A2026.05 | 26,745.9 | 13,956.9 | |
| Greedy-Laplacek=1002026.05 | 27,527 | 13,802.1 | |
| Last kk=5002026.05 | 32,131.9 | 17,366.6 | |
| Last kk=1002026.05 | 38,473.7 | 16,114.2 | |
| Subnet Diagonalk=5002026.05 | 48,953.1 | 21,940.2 | |
| MAP (no exploration)k=N/A2026.05 | 50,082.2 | 19,103.3 | |
| Subnet Diagonalk=1002026.05 | 60,640 | 20,071.5 |