Judea Pearl

来自集智百科 - 复杂系统|人工智能|复杂科学|复杂网络|自组织
Aceyuan讨论 | 贡献2022年6月20日 (一) 10:07的版本 (研究领域)
跳到导航 跳到搜索

基本信息

朱迪亚·珀尔
朱迪亚·珀尔
类别 信息
生日 1936年9月4日
出生地 以色列特拉维夫
国籍 美国
居住地 美国加利福尼亚州洛杉矶市
任职 加州大学洛杉矶分校,计算机科学教授
著名成就 贝叶斯网络

珀尔因果层次模型 PCH(Pearl Causal Hierarchy),又称因果之梯 The Ladder of Causation

主要研究方向 人工智能、因果推理和科学哲学
教育院校 以色列理工学院

纽瓦克工程学院(现新泽西理工学院)

新泽西州新不伦瑞克市罗格斯大学

纽约布鲁克林理工学院(现纽约大学理工学院)



朱迪亚·珀尔(Judea Pearl)——以色列裔美籍计算机科学家、哲学家,以倡导人工智能的概率方法和贝叶斯网络而闻名。他因发明了贝叶斯网络、定义复杂概率模型的数学形式以及这些模型中用于推理的主要算法而受到赞誉。这项工作不仅彻底改变了人工智能领域,而且成为许多其他工程和自然科学分支的重要工具。他后来创建了一个因果推理的数学框架,该框架对社会科学产生了重大影响。ACM授予Judea Pearl 2011年度图灵奖,以表彰他“通过发展概率和因果推理演算对人工智能做出的基础性贡献”。

他早在40多年前便通过贝叶斯网络的设计,使机器实现概率推理而在人工智能领域声名大噪,并被誉为“贝叶斯网络之父”,但近年却公开声称自己其实是人工智能社区的一名“叛徒”:离开了主流追逐、并且也是由他奠定重要理论基础和方法论的概率推理,而去追求一项更具挑战性的任务——因果推理。Judea Pearl 认为当今深度学习所有令人印象深刻的成就,都只不过是为了适应“曲线拟合(Curve fitting)”。而今,这也导致深度学习的研究员们困在了因果之梯的最底层——“关联”层次的问题窘境里。Judea Pearl 期望能掀起一场“因果革命”,采用因果推理模型,从因果而非单纯的数据关联角度去研究人工智能。

成长经历

求学

Judea Pearl 于1960年在海法的以色列理工学院获得电气工程学士学位。他于1961年在纽瓦克工程学院(现为新泽西理工学院)获得电气工程硕士学位。1965年,他在新泽西州新不伦瑞克市的罗格斯大学获得物理学硕士学位,同年,在纽约布鲁克林理工学院(现纽约大学理工学院)获得电气工程博士学位。他的博士学位论文是“超导记忆的涡旋理论”,“Pearl 涡旋(Pearl Vortex)”就是用来描述他所研究的超导电流的类型,这个词在物理学家中很流行。

工作

Pearl 曾在新泽西州普林斯顿的 RCA 研究实验室从事超导参数放大器和存储器件方面的工作,并在加利福尼亚州霍桑市的 Electronic Memory, Inc. 从事高级存储系统方面的工作。尽管当时他的工作聚焦在物理器件方面,Pearl 说从那时起他就对智能系统潜在应用充满向往。

学术

当磁性和超导存储器的工业研究因大规模半导体存储器的出现而减少时,Pearl 决定进入学术界以追求他对逻辑和推理的长期兴趣。1969 年,他加入加州大学洛杉矶分校,最初在工程系统系任教,并于1970年在新成立的计算机科学系获得终身教职。1976年晋升为正教授。1978年,他创立了认知系统实验室——这个名称强调了他对理解人类认知的愿望。

研究领域

搜索和启发式

Pearl 在计算机科学领域的声誉最初不是建立在概率推理(这在当时是一个有争议的话题)上,而是建立在组合搜索上。从1980年开始发表一系列期刊论文,最终于1984年 Pearl 出版了《启发式:计算机问题解决的智能搜索策略》一书。这项工作包括许多关于传统搜索算法的新结果,例如A*算法,以及在游戏算法方面,将人工智能研究提升到一个新的严谨和深度水平。它还提出了关于如何从宽松的问题定义中自动推导出可接受的启发式的新想法,这种方法导致了规划系统的巨大进步。

贝叶斯网络

Pearl 认为,对问题进行合理的概率分析会给出直观正确的结果,即使在基于规则的系统行为不正确的情况下也是如此。一个这样的案例与因果推理(从原因到结果)和诊断推理的能力有关(从结果到原因)。“如果使用诊断规则,则无法进行预测,如果使用预测规则,则无法进行诊断推理,如果同时使用两者,则会遇到正反馈不稳定性,这是我们在概率论中从未遇到过的。” 另一个案例涉及“解释消失”现象,即当观察到给定结果时,对导致结果的任何原因的相信程度会增加,但当发现其他原因也能导致观察到的结果时,对之前原因的相信程度就会降低。基于规则的系统不能表现出“解释消失”现象,而它在概率分析中会自动发生。

Pearl 意识到条件独立的概念将是构建具有多项式多参数的复杂概率模型和组织分布式概率计算的关键。论文“Reverend Bayes on Inference Engines: A Distributed Hierarchical Approach”[2]介绍了由有向无环图定义的概率模型,并推导出了一种精确的、分布式的、异步的、线性时间的树推理算法——我们现在称之为信念传播的算法,turbocodes的基础。随后,Pearl 有一段非凡的创意产出,发表了 50 多篇论文,涵盖一般图的精确推理、使用马尔可夫链蒙特卡罗的近似推理算法、条件独立属性、学习算法等,直到 1988 年出版了《智能系统中的概率推理》。这部具有里程碑意义的著作将珀尔的哲学、他的人类认知理论和他所有的技术材料结合成一个有说服力的整体这引发了人工智能领域的一场革命。在短短几年内,来自人工智能内部逻辑阵营和神经网络阵营的主要研究人员采用了一种概率(通常简称为现代)方法来研究人工智能。

Pearl 的贝叶斯网络为多元概率模型提供了句法和演算,就像乔治·布尔为逻辑模型提供句法和演算一样。与贝叶斯网络相关的理论和算法问题是机器学习和统计学现代研究议程的重要组成部分,它们的使用也渗透到其他领域,如自然语言处理、计算机视觉、机器人技术、计算生物学和认知科学。截至2012年,已经出现了大约 50,000 篇以贝叶斯网络为主要关注点的出版物。

因果关系

奖项与成就

多年来,他因在人工智能、人类推理和科学哲学领域做出重大贡献而享誉国际。获得近50项各类奖项(http://bayes.cs.ucla.edu/jp_home.html

以下为重要奖项:

2001 年,他因提出科学哲学方面的最佳著作而获得伦敦经济学院授予的拉科塔斯奖。

2003 年,他获得了 ACM 艾伦纽厄尔奖。

2006 年,获得了 Civic Venture 的首届目的奖,该奖项旨在表彰 60 岁及以上在解决社区和国家问题方面表现出非凡远见的个人。

2008 年,富兰克林研究所授予他本杰明富兰克林计算机和认知科学奖章。

2011 年,他因对人类认知理论基础的贡献而获得大卫 E. Rumelhart 奖。他的母校授予他哈维科学技术奖。

2011 年,他获得了ACM的图灵奖,这是计算机工程领域的最高荣誉,以表彰他“通过开发用于概率和因果推理的微积分对人工智能的根本贡献”。

主要文章及著作

他就人工智能的各个主题发表了近500篇科学论文(http://bayes.cs.ucla.edu/jp_home.html)。此外,他在上述感兴趣的领域共出版五本著作:

1984 年,《启发法》 Heuristics, Addison-Wesley, 1984

1988 年,《智能系统中的概率推理:合理推断网络》 Probabilistic Reasoning in Intelligent Systems, Morgan-Kaufmann, 1988

2009 年,《因果关系:模型、论证、推理》 Causality: Models, Reasoning, and Inference, Cambridge University Press, 2000; 2nd edition, 2009.

2016 年,《统计因果推理入门》 Causal Inference in Statistics: A Primer, (with Madelyn Glymour and Nicholas P. Jewell) Wiley, 2016.

2018 年,《为什么:关于因果关系的新科学》 The Book of Why: The New Science of Cause and Effect (with Dana Mackenzie), New York: Basic Books, May 2018

个人生活

他和露丝结婚。这对夫妇生下了三个孩子,其中包括在巴基斯坦被基地组织武装分子绑架并杀害的记者丹尼尔·珀尔。这位以色列裔美国计算机科学家和哲学家的儿子、华尔街日报记者丹尼尔·珀尔(Daniel Pearl)于 2002 年在巴基斯坦被激进分子绑架并杀害。

在他的记者儿子去世后,他与家人一起创立了丹尼尔·珀尔基金会。该组织的主要目的是促进诚实的报道和东西方的理解。它还旨在达到犹太人和穆斯林之间的理解水平。该组织在 2002 年和 2003 年同时获得了两个奖项。

研究领域

(R-513):  [pdf]

S. Mueller and J. Pearl "Personalized Decision Making -- A Conceptual Introduction,"

UCLA Cognitive Systems Laboratory, Technical Report (R-513), April 2022.


(R-513):  [pdf]

S. Mueller and J. Pearl "Personalized Decision Making -- A Conceptual Introduction,"

UCLA Cognitive Systems Laboratory, Technical Report (R-513), April 2022.

(R-511):  [pdf] [bib]  

A. Li and J. Pearl "Bounds on Causal Effects and Application to High Dimensional Data,"

UCLA Cognitive Systems Laboratory, Technical Report (R-511), March 2022.

In Proceedings of the Thirty-Sixth AAAI Conference on Artificial Intelligence (AAAI-22).

(R-510):  [pdf] [bib]  

A. Li and J. Pearl "Unit Selection with Causal Diagram,"

UCLA Cognitive Systems Laboratory, Technical Report (R-510), March 2022.

In Proceedings of the Thirty-Sixth AAAI Conference on Artificial Intelligence (AAAI-22).

(R-509):  [pdf] [bib]  

A. Forney and S. Mueller "Causal Inference in AI Education: A Primer,"

UCLA Cognitive Systems Laboratory, Technical Report (R-509), June 2022.

Forthcoming, Journal of Causal Inference.

(R-508):  [pdf] [bib]  

C. Cinelli "Transparent and Robust Causal Inferences in the Social and Health Sciences,"

UCLA Cognitive Systems Laboratory, Technical Report (R-508), July 2021.

Ph.D. Thesis

(R-507):  [pdf] [bib]  

A. Li, "Unit Selection Based on Counterfactual Logic,"

UCLA Cognitive Systems Laboratory, Technical Report (R-507), June 2021.

Ph.D. Thesis

(R-506):  [pdf] [bib]  

S. Mueller, "Estimating Individualized Causes of Effects by Leveraging Population Data,"

UCLA Cognitive Systems Laboratory, Technical Report (R-506), June 2021.

Master's Thesis

(R-505):  [pdf] [bib]

S. Mueller, A. Li, and J. Pearl "Causes of effects: Learning individual responses from population data,"

UCLA Cognitive Systems Laboratory, Technical Report (R-505), Revised May 2022.

Forthcoming, Proceedings of IJCAI-2022. 5br>

(R-505-Supplemental):  [Supplemental]

(R-504):  [pdf] [bib]

C. Zhang, C. Cinelli, B. Chen, and J. Pearl "Exploiting equality constraints in causal inference,"

UCLA Cognitive Systems Laboratory, Technical Report (R-504), April 2021.

Proceedings of the 24th International Conference on Artificial Intelligence and Statistics (AISTATS), San Diego, California, USA. PMLR: Volume 130, 1630-1638, April 2021.

(R-504-Supplemental):  [Supplemental]

(R-503):  [pdf] [bib]

J. Pearl "Causally Colored Reflections on Leo Breiman's `Statistical Modeling: The Two Cultures' (2001),"

UCLA Cognitive Systems Laboratory, Technical Report (R-503), March 2021.

Observational Studies, Vol. 7.1:187-190, 2021.

(R-502):  [pdf] [bib]

J. Pearl "Radical Empiricism and Machine Learning Research,"

UCLA Cognitive Systems Laboratory, Technical Report (R-502), May 2021.

Journal of Causal Inference, 9:78–82, 2021.

(R-501):  [pdf] [bib]

J. Pearl "Causal, Casual, and Curious (2013-2020): A collage in the art of causal reasoning,"

UCLA Cognitive Systems Laboratory, Technical Report (R-501), October 2020.

(R-493):  [pdf] [bib]

C. Cinelli, A. Forney, and J. Pearl "A Crash Course in Good and Bad Controls,"

UCLA Cognitive Systems Laboratory, Technical Report (R-493), Revised, March 2022.

Forthcoming, Journal Sociological Methods and Research.

(R-492):  [pdf] [bib]

C. Cinelli and J. Pearl "Generalizing experimental results by leveraging knowledge of mechanisms,"

UCLA Cognitive Systems Laboratory, Technical Report (R-492), September 2020.

European Journal of Epidemiology, 36:149--164, 2021. URL https://doi.org/10.1007/s10654-020-00687-4.

(R-491-L):  [pdf] [bib]

C. Zhang, B. Chen, and J. Pearl "A Simultaneous Discover-Identify Approach to Causal Inference in Linear Models,"

UCLA Cognitive Systems Laboratory, Technical Report (R-491-L), February 2020.

Extended version of paper in Proceedings of the Thirty-fourth AAAI Conference on Artificial Intelligence (AAAI-2020), 34(6): 10318--10325, Palo Alto, CA: AAAI Press, 2020.

(R-489):  [pdf] [bib]

J. Pearl "The Limitations of Opaque Learning Machines,"

UCLA Cognitive Systems Laboratory, Technical Report (R-489), May 2019.

Chapter 2 in John Brockman (Ed.), Possible Minds: 25 Ways of Looking at AI, Penguin Press, 2019.

(R-488):  [pdf] [bib]

A. Li and J. Pearl "Unit Selection Based on Counterfactual Logic,"

UCLA Cognitive Systems Laboratory, Technical Report (R-488), June 2019.

In Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence (IJCAI-19), 1793-1799, 2019.

(R-488-Supplemental):  [Supplemental]

(R-487):  [pdf] [bib]

J. Pearl and Co-authored by D. Mackenzie, "Telling and re-telling history: The case for a whiggish account of the history of causation,"

UCLA Cognitive Systems Laboratory, Technical Report (R-487), March 2019.

(R-486):  [pdf] [bib]

J. Pearl, "On the interpretation of do(x),"

UCLA Cognitive Systems Laboratory, Technical Report (R-486), February 2019.

Journal of Causal Inference, Causal, Casual, and Curious Section, 7(1), online, March 2019.

(R-485):  [pdf] [bib]

J. Pearl, "Causal and counterfactual inference,"

UCLA Cognitive Systems Laboratory, Technical Report (R-485), December 2021.

In Markus Knauff and Wolfgang Spohn (Eds.), The Handbook of Rationality, Section 7.1, pp. 427-438, The MIT Press, 2021.

(R-484):  [pdf] [bib]

J. Pearl, "Sufficient Causes: On Oxygen, Matches, and Fires,"

UCLA Cognitive Systems Laboratory, Technical Report (R-484), September 2019.

Journal of Causal Inference, Causal, Casual, and Curious Section, AOP, https://doi.org/10.1515/jci-2019-0026, September 2019.

(R-483):  [pdf] [bib]

J. Pearl, "Does Obesity Shorten Life? Or is it the Soda? On Non-manipulable Causes,"

UCLA Cognitive Systems Laboratory, Technical Report (R-483), August 2018.

Journal of Causal Inference, Causal, Casual, and Curious Section, 6(2), online, September 2018.

(R-482):  [pdf] [bib]

C. Cinelli, D. Kumor, B. Chen, J. Pearl, and E. Bareinboim

"Sensitivity Analysis of Linear Structural Causal Models,"

UCLA Cognitive Systems Laboratory, Technical Report (R-482), June 2019.

Proceedings of the 36th International Conference on Machine Learning, PMLR 97, 1252-1261, 2019.

(R-481):  [pdf] [bib]

J. Pearl, "The Seven Tools of Causal Inference with Reflections on Machine Learning,"

UCLA Cognitive Systems Laboratory, Technical Report (R-481), July 2018.

Communications of ACM, 62(3): 54-60, March 2019

(R-480):  [pdf] [bib]

K. Mohan, F. Thoemmes, J. Pearl, "Estimation with Incomplete Data: The Linear Case,"

UCLA Cognitive Systems Laboratory, Technical Report (R-480), May 2018.

Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence (IJCAI-18), 5082-5088, 2018.

(R-479):  [pdf] [bib]

C. Cinelli and J. Pearl, "RE: A Practical Example Demonstrating the Utility of Single-world Intervention Graphs,"

UCLA Cognitive Systems Laboratory, Technical Report (R-479), April 2018.

Journal of Epidemiology, 29(6): e50--e51, November 2018.

(R-478):  [pdf] [bib]

J. Pearl and E. Bareinboim, "A note on `Generalizability of Study Results',"

UCLA Cognitive Systems Laboratory, Technical Report (R-478), April 2018.

Epidemiology, 30(2):186--188, March 2019.

(R-477):  [pdf] [bib]

J. Pearl, "Challenging the Hegemony of Randomized Controlled Trials: Comments on Deaton and Cartwright,"

UCLA Cognitive Systems Laboratory, Technical Report (R-477), April 2018.

Social Science and Medicine, published online, April 2018.

(R-476):  [pdf] [bib]

J. Pearl, "A Personal Journey into Bayesian Networks,"

UCLA Cognitive Systems Laboratory, Technical Report (R-476), May 2018.

(R-475):  [pdf] [bib]

J. Pearl, "Theoretical Impediments to Machine Learning with Seven Sparks from the Causal Revolution"

UCLA Cognitive Systems Laboratory, Technical Report (R-475), July 2018.

Paper supporting Keynote Talk WSDM'18, Proceedings of the Eleventh ACM International Conference on Web Search and Data Mining, DOI: http://dx.doi.org/10.1145/3159652.3160601, February 2018.

(R-474):  [pdf] [bib]

J. Pearl, "Comments on `The Tale Wagged by the DAG'"

UCLA Cognitive Systems Laboratory, Technical Report (R-474), January 2018.

International Journal of Epidemiology, 47(3):1002-1004, 2018.

(R-473):  [pdf] [bib]

K. Mohan and J. Pearl, "Graphical Models for Processing Missing Data"

UCLA Cognitive Systems Laboratory, Technical Report (R-473-L), June 2019.

Journal of American Statistical Association (JASA). Online March 2021 (https://doi.org/10.1080/01621459.2021.1874961).

(R-472):  [pdf] [bib]

J. Pearl, "What is Gained from Past Learning"

UCLA Cognitive Systems Laboratory, Technical Report (R-472), March 2018.

Journal of Causal Inference, Causal, Casual, and Curious Section, 6(1), Article 20180005, March 2018. https://doi.org/10.1515/jci-2018-0005

(R-471):  [pdf] [bib]

A. Forney, J. Pearl, and E. Bareinboim, "Counterfactual Data-Fusion for Online Reinforcement Learners"

UCLA Cognitive Systems Laboratory, Technical Report (R-471), June 2017.

Presented at the Transfer in Reinforcement Learning workshop at AAMAS-2017.

Proceedings of the 34th International Conference on Machine Learning, PMLR 70:1156-1164, 2017.

(R-470):  [pdf] [bib]

J. Pearl, "The Eight Pillars of Causal Wisdom"

UCLA Cognitive Systems Laboratory, Technical Report (R-470), April 2017.

(R-469):

J. Pearl, "A Personal Journey into Bayesian Networks,"

UCLA Cognitive Systems Laboratory, Technical Report (R-476), May 2018.

(R-466):  [pdf] [bib]

J. Pearl "The Sure-Thing Principle"

UCLA Cognitive Systems Laboratory, Technical Report (R-466), February 2016.

Journal of Causal Inference, Causal, Casual, and Curious Section, 4(1):81-86, March 2016.

(R-461):  [pdf] [bib]

B. Chen, J. Pearl, and E. Bareinboim, "Incorporating Knowledge into Structural Equation Models using Auxiliary Variables"

UCLA Cognitive Systems Laboratory, Technical Report (R-461), July 2016.

In S. Kambhampati (Ed.), Proceedings of the 25 International Joint Conference on Artificial Intelligence (IJCAI), Palo Alto: AAAI Press, 3577-3583, 2016.

(R-461-L):  [pdf]

B. Chen, J. Pearl, and E. Bareinboim, "Incorporating Knowledge into Structural Equation Models using Auxiliary Variables"

UCLA Cognitive Systems Laboratory, Technical Report (R-461-L), April 2016.

Extended version of paper in S. Kambhampati (Ed.), Proceedings of the 25 International Joint Conference on Artificial Intelligence (IJCAI), Palo Alto: AAAI Press, 3577-3583, 2016.

(R-460):  [pdf] [bib]

E. Bareinboim, Andrew Forney, and J. Pearl, "Bandits with Unobserved Confounders: A Causal Approach"

UCLA Cognitive Systems Laboratory, Technical Report (R-460), November 2015.

In C. Cortes, N.D. Lawrence, D.D. Lee, M. Sugiyama, and R. Garnett (Eds.), Neural Information Processing Systems (NIPS) Conference, Advances in Neural Information Processing Systems 28, Curran Associates, Inc., pp. 1342-1350, 2015.

(R-459):  [pdf] [bib]

J. Pearl, "A Linear `Microscope' for Interventions and Counterfactuals"

UCLA Cognitive Systems Laboratory, Technical Report (R-459), March 2017.

Journal of Causal Inference, Causal, Casual, and Curious Section, published online 5(1):1-15, March 2017.

参考文献

[1]J. Pearl, "A Personal Journey into Bayesian Networks,"

UCLA Cognitive Systems Laboratory, Technical Report (R-476), May 2018.

编者推荐

集智俱乐部推文推荐

统计学权威盘点过去50年最重要的统计学思想,因果推理、bootstrap等上榜,Judea Pearl点赞 | 集智俱乐部

福利 | 因果推断会是下一个AI热潮吗?Judea Pearl《因果论》重磅上市!

Stephen Wolfram专访Judea Pearl:从贝叶斯网络到元胞自动机 | 集智俱乐部

说明

J. Pearl 发表很多论文,是困难的 去编写问题 从Pearl 的论文 使用自己的语言。因此,我采用多轮次去编写。每个轮次编写1~2个问题。更多问题将编写。

  • 如何编写问题? 请参考“V形图