Always-On Experimentation

研究问题: 摘要称，生成式 AI 加快了新处理（如新药、线上营销活动）的产生与部署速度，实验平台因此持续运行、处理可随时加入或移除；论文研究在此“Always-On”设定下，如何对每个处理做出接受或拒绝的决策，同时控制错误发现率。
方法与思路: 作者提出基于 testing-by-betting 框架的序贯检验：为每个处理的平均处理效应构造检验超鞅，在任意停时与“可预测”的处理时间表下实现时间一致的第一类错误控制，并称该构造在几乎必然意义下具有增长率最优性。
材料陈述的贡献: 摘要明确陈述的贡献是：形式化“Always-On”实验设定（处理可动态生成、加入与移除），并给出在任意停时和可预测处理安排下具备时间一致 Type-I 错误控制、且增长率最优的序贯检验构造。
阅读关注点: 阅读全文时应核对：平均处理效应的具体识别假设与目标参数定义；“可预测”处理时间表的确切含义与适用范围；时间一致错误控制与 FDR 控制之间的衔接方式；增长率最优性的形式化陈述与证明条件；是否包含模拟或实证校准。
阅读边界: 仅凭摘要无法判断识别假设是否成立、定理条件是否可满足、最优性结论的适用范围，以及是否存在数值或实证验证；摘要未说明具体数据、样本或应用场景。该来源为 arXiv 预印本，未经同行评审验证，不能据此声称已被期刊接收。

DeepSeek Flash · 2026-10-01

原文摘要

Generative AI has dramatically accelerated the rate at which new treatments---from novel pharmaceuticals to online marketing campaigns---can be conceived and deployed. As a result, modern experimentation platforms often run continuously, with treatments added as they are ready and removed when they underperform. We formalize this "Always-On" experimental setting, in which treatments can be dynamically generated, added to, and removed from a running experiment, and study the statistical problem of deciding whether to accept or reject each treatment while controlling for the false discovery rate. We develop sequential tests that achieve time-uniform Type-I error control under arbitrary stopping times and "predictable" treatment schedules. Our approach builds on the testing-by-betting framework: we construct test supermartingales for testing the average treatment effect of each treatment, and show that the construction of these test supermartingales is growth-rate optimal in an almost-sure sense.

摘要来源：arXiv

笔记与分组

同方向的基础文献

相关领域论文

共同方法：因果推断、检验与推断、多重检验与同时推断 · 共同方向：因果推断

Debiased inference for proximal dose-response function

Daeyoung Ham · Sihan Wu · Yifan Cui首次提交 2026-08-01 · 修订 2026-10-07 · v2arXiv预印本

We study nonparametric inference for a continuous-treatment causal dose-response curve and its first derivative under unmeasured confounding, using treatment- and outcome-inducing proxies. We establish a proximal doubly robust pseudo-outcome whose conditional mean given treatment equals the dose-response curve whenever either bridge function is correctly specified. Their (smoothed) efficient influence function representations lead to cross-fitted local-polynomial estimators…

微观 · 因果推断核、筛与局部多项式半参数与效率

关联记录被引 0 次 ↗OpenAlex · 核验 2026-10-01

阅读原文 ↗BibTeX ↓

共同方法：因果推断、检验与推断、多重检验与同时推断 · 共同方向：因果推断

Longitudinal causal inference under informative missingness using the self-censoring model

Iván Díaz首次提交 2026-10-06 · 修订 2026-10-06 · v1arXiv预印本

自删失模型下信息性缺失的纵向因果推断AI 解读

Longitudinal observational data are often subject to informative missingness. For example, in healthcare research using electronic health records, covariates and outcomes are measured only when patients interact with the healthcare system. Whether these data are observed usually depends on the unobserved variable values themselves, introducing a common form of data missing not at random. We develop identification and estimation methods for the effects of longitudinal…

微观 · 因果推断面板数据机器学习与 DML

被引：暂无数据

AI 解读阅读原文 ↗BibTeX ↓

共同方法：因果推断、检验与推断、多重检验与同时推断 · 共同方向：因果推断

Interpretable Discovery from Unstructured Data: A High-Dimensional Approach

Jacob Carlson首次提交 2025-11-03 · 修订 2026-10-02 · v5arXiv预印本

We propose an automatic, general-purpose framework for making discoveries from unstructured data (e.g., text data from open-ended surveys of economic beliefs). The framework leverages recent methods from the literature on AI interpretability to transform unstructured datasets into high-dimensional, structured datasets of interpretable concept measurements; specifies concept-level parameters and null hypotheses based on this transformed dataset; tests these hypotheses using…

微观 · 因果推断高维与机器学习多重检验与同时推断

关联记录被引 0 次 ↗OpenAlex · 核验 2026-10-07

阅读原文 ↗BibTeX ↓

共同方法：因果推断、检验与推断、多重检验与同时推断 · 共同方向：因果推断

Debiased and Simultaneous Inference for Heterogeneous Factorial Effect Modifiers via Residualized Walsh-Hadamard Scores

Tomoshige Nakamura · Ryo Emoto首次提交 2026-09-29 · 修订 2026-09-29 · v1arXiv预印本

基于残差化 Walsh–Hadamard 得分的异质析因效应修饰因子去偏与同时推断AI 解读

In factorial randomized experiments with $K$ binary treatment components, we study which baseline covariates modify component main effects and interactions. We define effect-modifier cells $(j,S)$, $j\in[p]$ and $\emptyset\neq S\subseteq[K]$, through the coefficients $θ^\ast_{jS}$ of population linear projections of conditional Walsh--Hadamard contrasts $τ_{S}$ onto standardized covariates. Using baseline-robust direct scores, we develop frequentist inference for all $p(2^K-1)$…

微观 · 因果推断多重检验与同时推断正则化与变量选择

关联记录被引 0 次 ↗OpenAlex · 核验 2026-10-08

AI 解读阅读原文 ↗BibTeX ↓

共同方法：因果推断、检验与推断、多重检验与同时推断 · 共同方向：因果推断

Simultaneous Inference for Proximal Causal Learning with Weak Proxies

Pengyun Wang · Qi Huang · Jia Huang · Yutong Zhou首次提交 2026-05-10 · 修订 2026-09-25 · v2arXiv预印本

弱代理变量下近端因果学习的同时推断AI 解读

Proximal causal learning uses the treatment-inducing proxy and the outcome-inducing proxy through the outcome bridge and the treatment bridge to provide a way to address unobserved confounding in observational studies. This paper studies simultaneous inference for the marginal counterfactual distribution and quantile treatment effect under weak proxies. First, we propose a measure of weak-proxy strength and tightens the population-bias bound for the counterfactual CDF…

微观 · 因果推断多重检验与同时推断微观 · 分位数与分布效应

关联记录被引 0 次 ↗OpenAlex · 核验 2026-10-01

AI 解读阅读原文 ↗BibTeX ↓

共同方法：因果推断、检验与推断、多重检验与同时推断 · 共同方向：因果推断

Statistical Inference for Causal Discovery under Selection and Latent Variables via Single-Target Interventions

Xiaotian Hou · Kwangmoon Park · Hongzhe Li首次提交 2026-09-23 · 修订 2026-09-23 · v1arXiv预印本

潜变量与选择偏差下基于单目标干预的因果发现推断AI 解读

Causal discovery from observational and interventional data becomes challenging in the presence of latent confounding and selection bias, where causal structure is no longer adequately represented by directed acyclic graphs over observed variables. Existing model-free methods often rely on an exponential number of conditional independence tests and provide limited uncertainty quantification in high-dimensional settings. We develop a model-free and constraint-query optimal…

微观 · 因果推断高维与机器学习多重检验与同时推断

关联记录被引 0 次 ↗OpenAlex · 核验 2026-10-01

AI 解读阅读原文 ↗BibTeX ↓