Safe Exploration in Bandits on Trap Bandit alpha_DGP=0.01
0Catastrophe Rateinfra bayesian
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| infra bayesianalpha_prior=n/a, Exploration strategy=greedy2026.05 | 0 | 39.6 | 40 | |
| bayes greedyalpha_prior=0.01, Exploration strategy=greedy2026.05 | 1.5 | 0.4 | 5.6 | |
| bayes thompsonalpha_prior=0.01, Exploration strategy=Thompson sampling2026.05 | 1.5 | 1.6 | 7.6 |