方法对比
并排查看您选择的方法；存在差异的行会高亮显示。
| 多臂老虎机 (UCB, Thompson Sampling)× | 顺序/分组顺序试验设计× | |
|---|---|---|
| 领域 | 实验设计 | 实验设计 |
| 方法族 | Hypothesis test | Hypothesis test |
| 起源年份≠ | 1952 | 1979 |
| 提出者≠ | Robbins (1952); UCB1 by Auer et al. (2002); Thompson sampling by Thompson (1933) | O'Brien & Fleming; Pocock; Lan & DeMets |
| 类型≠ | Sequential decision / bandit algorithm | Adaptive stopping trial design |
| 开创性文献≠ | Auer, P., Cesa-Bianchi, N., & Fischer, P. (2002). Finite-Time Analysis of the Multiarmed Bandit Problem. Machine Learning, 47(2–3), 235–256. DOI ↗ | O'Brien, P.C. & Fleming, T.R. (1979). A Multiple Testing Procedure for Clinical Trials. Biometrics, 35(3), 549–556. DOI ↗ |
| 别名≠ | MAB, bandit algorithm, UCB1, Thompson sampling | group sequential design, adaptive stopping design, Ardışık Deneme Tasarımı (Sequential / Group Sequential) |
| 相关≠ | 4 | 3 |
| 摘要≠ | The multi-armed bandit (MAB) is an adaptive experimental framework that allocates trials sequentially across competing arms to minimise cumulative regret while simultaneously learning which arm performs best. Formalised by Robbins in 1952 and given finite-time guarantees by Auer et al. (2002), it balances exploration of uncertain options against exploitation of currently known best options — outperforming classical A/B testing whenever early stopping or cost-sensitive allocation matters. | Sequential and group sequential trial designs allow a study to be stopped early — or continued — based on interim analyses conducted as data accumulate. The core framework was formalised by O'Brien and Fleming in 1979 and extended by Lan and DeMets's alpha-spending approach, and it controls the overall Type I error rate across all planned looks by pre-specifying both efficacy and futility boundaries before enrolment begins. |
| ScholarGate数据集 ↗ |
|
|