Causality

Judea Pearl

出版时间

2009-09-13

ISBN

9780521895606

评分

★★★★★
书籍介绍

Written by one of the preeminent researchers in the field, this book provides a comprehensive exposition of modern analysis of causation. It shows how causality has grown from a nebulous concept into a mathematical theory with significant applications in the fields of statistics, artificial intelligence, economics, philosophy, cognitive science, and the health and social sciences. Judea Pearl presents and unifies the probabilistic, manipulative, counterfactual, and structural approaches to causation and devises simple mathematical tools for studying the relationships between causal connections and statistical associations. Cited in more than 2,100 scientific publications, it continues to liberate scientists from the traditional molds of statistical thinking. In this revised edition, Judea Pearl elucidates thorny issues, answers readers' questions, and offers a panoramic view of recent advances in this field of research. Causality will be of interest to students and professionals in a wide variety of fields. Dr Judea Pearl has received the 2011 Rumelhart Prize for his leading research in Artificial Intelligence (AI) and systems from The Cognitive Science Society.

AI导读
核心看点
  • 本书系统构建了因果推断的数学理论框架,统一了概率论、结构因果模型与反事实推理,提供将统计相关性转化为因果关系的严格数学工具,彻底革新了传统统计学思维。
  • 深入解析贝叶斯网络与有向无环图在因果建模中的应用,阐明如何通过图模型识别混杂因子、处理辛普森悖论,并基于结构方程模型进行合法的因果效应计算与推断。
  • 探讨因果关系的哲学基础与认知逻辑,强调数据本身无法自动产生因果结论,必须结合对数据生成过程的语义理解与合理假设,揭示了人类认知中因果解释的深层机制。
适合谁读
  • 具备扎实概率论、统计学及图论基础的科研人员与研究生,特别是从事人工智能、机器学习、因果推断算法研究的专业人士,需能接受高难度数学推导。
  • 对科学方法论、认知科学及哲学感兴趣的研究者,希望深入理解因果逻辑本质,批判性审视传统统计关联分析的局限性,探索知识表示与推理的底层原理。
  • 经济学、社会学、公共卫生等领域的学者,若需进行严谨的政策评估或机制分析,且愿意投入大量时间攻克数学壁垒,以获取最权威的理论指导与方法论支持。
读前提醒
  • 本书数学密度极高,非科普读物,切勿期望轻松阅读。建议先掌握概率论基础,若数学背景薄弱,请先阅读作者通俗著作《为什么》或放弃,避免陷入公式泥潭导致挫败感。
  • 不要试图线性通读全书,建议重点研读第2、3、4、7章核心内容,结合具体案例理解图模型操作。对于晦涩部分,可参考其他通俗解释或跳过,保持对整体框架的理解。
  • 阅读时需警惕将统计相关性误认为因果关系的思维惯性,书中强调假设的重要性,读者应时刻反思数据背后的生成机制,避免陷入‘唯数据论’的误区,注重逻辑推导的合法性。
读者共识
  • 作为因果推断领域的奠基之作,本书学术地位极高,是理解现代因果理论不可绕过的经典,但因其极高的专业门槛和晦涩的表达,被广泛认为不适合初学者或非专业人士阅读。
  • 读者普遍认为本书不仅是技术手册,更蕴含深刻的哲学思考,关于因果本质、认知推理的讨论极具启发性,但部分章节存在啰嗦、重复或逻辑跳跃问题,阅读体验较为痛苦。
  • 尽管存在阅读障碍,但掌握其核心思想后能显著提升科研思维与逻辑严谨性,许多读者在历经艰难阅读后表示‘醍醐灌顶’,认为其提供的理论工具对纠正错误统计实践具有决定性价值。

本导读基于书籍简介、目录、原文摘录、短评和书评生成,不等同于全文精读。

精彩摘录
  • "「他们很快意识到,确定X和Y之间因果关系方向的关键在于“存在第三个变量Z,与Y相关,与X无关」"
  • "「无向图有时称为马尔可夫网络(Markov networks)(Pearl,。1988b),主要用于表示对称的空间关系(Isham,1981;Cox and Wermuth,1996;Lauritzen,1996)。有向图,尤其是DAG,用于表示因果关系或时间关系(Lauritzen,1982;Wermuth and Lauritzen,1983;Kiiveri et al.,1984)。这种图称为贝叶斯网络(Bayesian networks)」"
  • "However, this corresponds to a general pattern of causal relationships: observations on a common consequence of two independent causes tend to render those causes dependent, because information about one of the causes tends to make the other more or less likely, given that the consequence has occurr"
  • "In view of this stability, it is no wonder that people prefer to encode knowledge in causal rather than probabilistic structures. Probabilistic relationships, such as marginal and conditional independencies, maybe helpful in hypothesizing initial causal structures from uncontrolled observations. How"
  • "「“鲁宾因果模型”或者“潜在结果模型”,其基本内容是,假设因素A和因素B都只有两个状态,为了考察因素A对于因素B的因果效应,在固定其他因素Z的前提下,设置A的两个状态A0、A1,分别计算对于B的状态影响,即计算E(B|A0,Z) = Ez{B(0)}和E(B|A1,Z) = Ez{B(1)},它们的差Ez{B(0)} - EzB(1)}def ACE(A → B)被定义为A对于B的平均因果效应(由于在Z固定的情况下,实际能够观察到的数据要么是A=A0,要么是A=A1,不可能同时观察到两种状态,因此有一个状态是现实中不存在的,是假设的,这个假设状态所引起的B的变化称为“潜在结果」"
  • "「数据的适用性是验证因果理论的不充分标准」"
  • "「但是因果分析的假定所需要的数据支持却与数据数量无关,这一点与先验假设有很大的不同。在因果分析中,一个(或一组)合理的假定会推断出相应的因果关系,如果不同意这个假定,自然也就否认因果关系成立。不同的假定会得到不同的因果结论,当然,越是简单和直观的假定,越有可能推出符合实际问题的因果结论。所以,只有数据本身是做不了因果分析的,还需要知道数据生成的过程(即数据的语义),才能做出合理的假定」"
  • "「在人类话语中,因果解释满足两个期望:时间和统计。时间方面用“原因应先于结果”这一认知来表示,统计方面则期望有一个完全的因果解释来涵盖它的各种效应(即,使效应有条件的独立);不能涵盖其效应的解释被认为是“不完整的”,剩余的相关性部分被认为是“虚假的”或“无法解释的”」"
作者简介
Judea Pearl is professor of computer science and statistics at the University of California, Los Angeles, where he directs the Cognitive Systems Laboratory and conducts research in artificial intelligence, human reasoning, and philosophy of science. The author of Heuristics and Probabilistic Reasoning, he is a member of the National Academy of Engineering and a Founding Fellow of the American Association for Artificial Intelligence. Dr Pearl is the recipient of the IJCAI Research Excellence Award for 1999, the London School of Economics Lakatos Award for 2001, and the ACM Alan Newell Award for 2004. In 2008, he received the Franklin Medal for computer and cognitive science from the Franklin Institute.
目录
1. Introduction to probabilities, graphs, and causal models;
2. A theory of inferred causation;
3. Causal diagrams and the identification of causal effects;
4. Actions, plans, and direct effects;
5. Causality and structural models in social science and economics;

显示全部
用户评论
讨论小组一起看了第一章。
看得出来大佬想探讨众多因果相关定义背后的直观背景,如条件独立性等等(涉及bayes的概率理解),但也因此丧失了部分严格性。对sem和反事实的内在原理阐述的极为清晰~
似醍醐灌顶
大佬的教科书,不如他和人合写的那本清晰,这本有点啰嗦,当然也更全更细。
与其说是统计学/人工智能的书,不如说是哲学书。把这本书当作单纯介绍因果推断理论的专著是对Pearl思想精髓的断章取义,Pearl关注的不只是因果的问题,而且是关于人类认知推理的大问题。
第一遍读,有挺多地方没读懂,大概弄明白了因果这套东西是做什么的,以及怎么做的,大概要边实践边思考了才能理解更深了
收藏