The Cost of Sequential Adaptation and the Lower Bound for Mean Squared Error

Effect of Interim Adaptations in Group Sequential Designs

This manuscript investigates unconditional and conditional-on-stopping maximum likelihood estimators (MLEs), information measures and information loss associated with conditioning in group sequential designs (GSDs). The possibility of early stopping brings truncation to the distributional form of MLEs; sequentially, GSD decisions eliminate some events from the sample space. Multiple testing induces mixtures on the adapted sample space. Distributions of MLEs are mixtures of truncated distributions. Test statistics that are asymptotically normal without GSD, have asymptotic distributions, under …

Efficiency Requires Adaptation

The majority of historical designs are a priori in nature, where a priori indicates a design can be specified in advance of the experiment. The conventional wisdom is that the set of a priori designs is sufficient to produce efficient experiments. This work challenges this convention and finds that efficiency requires data dependent strategies. Specifically, in the context of a sequential experiment, where observations are accrued in a series of runs, an adaptive design is proposed that is guaranteed to be more efficient than any corresponding a priori design.

序贯决策中用于高效策略评估的最优治疗分配

Optimal Treatment Allocation for Efficient Policy Evaluation in Sequential Decision Making

针对序贯决策场景下的在线实验治疗效应精准估计问题，提出三种适配非/时变马尔可夫决策过程的最优分配策略，结合OPE方法验证有效性，理论证明其最优性并给出估计量均方误差上界。

使用拉丁超立方采样的分层分层域顺序估计

Sequential Estimation using Hierarchically Stratified Domains with Latin Hypercube Sampling

本文提出一种自适应顺序分层估计方法，结合拉丁超立方采样，通过分层细化降低方差，数值实验显示方差衰减可达N^{-2}。

贝叶斯反问题中代理建模的序贯设计

Sequential design for surrogate modeling in Bayesian inverse problems

本文针对贝叶斯反问题中的代理建模提出CSQ和IP-SUR两种序贯设计策略，利用SUR框架实现不确定性逐步减少，并通过多个测试案例验证其有效性。

概率图上用于可解释反事实公平性的序列条件传输

Sequential Conditional Transport on Probabilistic Graphs for Interpretable Counterfactual Fairness

本文结合因果图适配与最优传输，将Knothe重排和三角传输扩展至概率图模型，提出序列传输方法以探讨个体层面的反事实公平性，并通过数值实验验证其应用。