← 研踪 PaperTrailarXiv:2611.24988 · v1
Variance-Adaptive Upper Confidence Bounds for Multi-Armed Bandits with Kaplan--Meier Survival Estimation
首次提交 2026-11-18 13:42
版本更新 2026-11-18 13:42
雷达观测 2026-11-21 12:19
摘要
We study the stochastic multi-armed bandit problem in which the reward associated with each arm is the survival probability at a pre-specified time horizon, and the observed outcome is a right-censored time-to-event variable. Standard bandit algorithms face delayed and partial observations in this setting. Prior adaptations for the Kaplan--Meier (KM) estimator rely on Hoeffding-type bounds that ignore the inherent heteroscedasticity of continuous survival processes. To bridge this gap, we propose KM-UCB, a variance-adaptive policy that unites continuous-time survival analysis with discrete sequential decision-making. By leveraging martingale theory and continuous influence functions, we derive a fixed-cohort empirical Bernstein bound parameterized by Greenwood's variance formula. For realtime allocation, we retain the Greenwood index and use a survival bound from completed records to review the recommended arm. Under catch-up reporting, the policy has logarithmic expected regret with an arm-specific variance term in the exploration bound. We evaluate the original policy in a dual-layer delayed simulation and a retrospective replay of breast cancer trial data, and evaluate the correction in a separate paired simulation.
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v1 · 2026-11-18 13:42 · 最新观测版本
Variance-Adaptive Upper Confidence Bounds for Multi-Armed Bandits with Kaplan--Meier Survival Estimation
We study the stochastic multi-armed bandit problem in which the reward associated with each arm is the survival probability at a pre-specified time horizon, and the observed outcome is a right-censored time-to-event variable. Standard bandit algorithms face delayed and partial observations in this setting. Prior adaptations for the Kaplan--Meier (KM) estimator rely on Hoeffding-type bounds that ignore the inherent heteroscedasticity of continuous survival processes. To bridge this gap, we propose KM-UCB, a variance-adaptive policy that unites continuous-time survival analysis with discrete sequential decision-making. By leveraging martingale theory and continuous influence functions, we derive a fixed-cohort empirical Bernstein bound parameterized by Greenwood's variance formula. For realtime allocation, we retain the Greenwood index and use a survival bound from completed records to review the recommended arm. Under catch-up reporting, the policy has logarithmic expected regret with an arm-specific variance term in the exploration bound. We evaluate the original policy in a dual-layer delayed simulation and a retrospective replay of breast cancer trial data, and evaluate the correction in a separate paired simulation.
查看此版本解读 →查看此版本原文 ↗