阅读说明

本页整理公开书目信息，不生成学术质量评分。AI 解读根据题名及所注明的摘要或原文片段生成，未完成全文核验，可能有误；具体假设、证明与结论请核对原文。摘要保留来源文本，未经模型改写。

DOI 首次登记不等于接收或发表日期。部分更正、评论等条目没有独立摘要；缺少可靠材料时保留待补状态。

本页为预印本版本；arXiv 收录、作者填写的期刊信息以及作者既有发表经历，都不单独证明本篇已接收。作者发表记录按 OpenAlex 身份与完整姓名、本站覆盖期刊 ISSN 及正式卷期核对，只展示已核实例，档案仍可能误合并。

引用量采用 OpenAlex 单条关联记录，不合计多个记录；关联记录可能包含正式发表版本，不等于此版本单独引用。缺失不视为零，引用指标不代表论文质量。

关注作者按书目中的完整姓名匹配，同名和姓名变体需人工辨别。相关论文先按共享细分方向和具体方法排列，再参考通用方法与发表日期。

分类依据与纠错

AI 根据所提供的论文材料归类：摘要围绕条件独立检验、因果发现与因果结构的唯一识别展开。

machine learningcausalidentification

反馈分类问题

Causal de Finetti: On the Identification of Invariant Causal Structure in Exchangeable Data

Siyuan Guo · Viktor Tóth · Bernhard Schölkopf · Ferenc Huszár首次提交 2022-03-29 · 修订 2024-05-24 · v3arXiv预印本

因果 de Finetti：可交换数据中不变因果结构的识别AI 解读 ↓

微观 · 因果推断机器学习与 DML

关联记录被引 4 次 ↗OpenAlex · 核验 2026-09-30

阅读原文 ↗BibTeX ↓

正在核对关联版本…

DOI · 预印本使用 arXiv 编号

arXiv

作者的期刊发表依据

暂未核实关联作者的本站覆盖期刊正式发表记录。仍保留在“全部预印本”；资料缺失不代表没有发表经历。

arXiv：2203.15756 · v3

首次提交：2022-03-29 · 最近修订：2024-05-24

原站分类：stat.ML / cs.LG / math.ST / stat.ME

收录依据：统计方法、统计理论或金融计量分类，且题名或摘要含计量研究主题

下载 BibTeX 下载 RIS

代码、数据与补充材料

正在核对对应版本的资源…

关注这篇论文的作者

AI 解读 · 基于可核验材料

因果 de Finetti：可交换数据中不变因果结构的识别

研究问题: 独立同分布数据中的条件独立关系通常只能识别马尔可夫等价类；可交换数据提供的额外条件独立结构能否识别唯一的因果结构？
方法与思路: 作者提出因果 de Finetti 表示定理，将满足特定条件独立关系的可交换分布表示为独立因果机制生成过程，并据此设计条件独立检验算法。
材料陈述的贡献: 摘要称，在独立因果机制生成过程下，条件独立检验可识别唯一因果结构；论文同时提出发现算法，并展示其在多环境数据上的应用。
阅读关注点: 应核对可交换性的层级、非平凡条件独立要求、独立因果机制的定义，以及唯一识别结论如何转化为有限样本的可靠检验。
阅读边界: 依据所给 arXiv 预印本摘要，不能确认同行评审或期刊接收。摘要未列出完整定理条件，不能将唯一识别结论推广到任意可交换数据。

GPT-6.1 Sol · 高 · 2026-10-03

原文摘要

Constraint-based causal discovery methods leverage conditional independence tests to infer causal relationships in a wide variety of applications. Just as the majority of machine learning methods, existing work focuses on studying $\textit{independent and identically distributed}$ data. However, it is known that even with infinite i.i.d. $\$ data, constraint-based methods can only identify causal structures up to broad Markov equivalence classes, posing a fundamental limitation for causal discovery. In this work, we observe that exchangeable data contains richer conditional independence structure than i.i.d. $\$ data, and show how the richer structure can be leveraged for causal discovery. We first present causal de Finetti theorems, which state that exchangeable distributions with certain non-trivial conditional independences can always be represented as $\textit{independent causal mechanism (ICM)}$ generative processes. We then present our main identifiability theorem, which shows that given data from an ICM generative process, its unique causal structure can be identified through performing conditional independence tests. We finally develop a causal discovery algorithm and demonstrate its applicability to inferring causal relationships from multi-environment data. Our code and models are publicly available at: https://github.com/syguo96/Causal-de-FinettiConstraint-based causal discovery methods leverage conditional independence tests to infer causal relationships in a wide variety of applications. Just as the majority of machine learning methods, existing work focuses on studying $\textit{independent and identically distributed}$ data. However, it is known that even with infinite i.i.d.$\ $ data, constraint-based methods can only identify causal structures up to broad Markov equivalence classes, posing a fundamental limitation for causal discovery. In this work, we observe that exchangeable data contains richer conditional independence structure than i.i.d.$\ $ data, and show how the richer structure can be leveraged for causal discovery. We first present causal de Finetti theorems, which state that exchangeable distributions with certain non-trivial conditional independences can always be represented as $\textit{independent causal mechanism (ICM)}$ generative processes. We then present our main identifiability theorem, which shows that given data from an ICM generative process, its unique causal structure can be identified through performing conditional independence tests. We finally develop a causal discovery algorithm and demonstrate its applicability to inferring causal relationships from multi-environment data. Our code and models are publicly available at: https://github.com/syguo96/Causal-de-Finetti

摘要来源：arXiv

笔记与分组