集智百科 - 复杂系统|人工智能|复杂科学|复杂网络|自组织 - 用户贡献 [zh-cn]

因果发现

2022-07-10T13:24:38Z

Kunwuz：

=== 简介 ===

因果发现，即从纯观测数据中发现并获取因果关系，在近几十年来作为基础的数据分析方法在各个学科发挥着重要的作用<ref>Glymour, Clark, Kun Zhang, and Peter Spirtes. "Review of causal discovery methods based on graphical models." ''Frontiers in genetics'' 10 (2019): 524.</ref>。

几乎所有的科学都是关于识别因果关系和支配它们的法律或规律性的。它们的规律。自十七世纪现代科学开始以来，有两种方法来发现因果关系：（1）操纵和改变系统中的某一些的特征，以观察其他特征是否发生变化；（2）观察系统特征的变化，而不进行操作。这两种方法都在十七世纪大放异彩，当时它们是互相交织在一起，就像今天一样。伊万杰里斯塔-托里切利操纵了站在一个装有水银的管子里的角度和形状，而管子中水银的高度没有变化。帕斯卡尔让人把托里切利设计的压力计抬到山上，以证明水银的高度确实随海拔高度而变化。伽利略曾从观测时间序列中确定了（定性的）木卫二卫星的轨道。开普勒从行星观测中得出了他的三大定律。牛顿则通过对太阳系的观察和一个单一的实验得出的引力定律奠定了现代物理学的基础。现代分子生物学是一个实验课题，但生物学的基础，在达尔文的《物种起源》中，只有一个实验，即种子的漂移。

发现因果关系的传统方法是使用干预措施或随机实验，这在很多情况下过于昂贵，过于耗时，甚至不可能。因此，通过分析纯粹的观察性数据来揭示因果信息，即所谓的因果发现，已经引起了人们的关注（Spirtes et al, 2000<ref>Spirtes, Peter, et al. ''Causation, prediction, and search''. MIT press, 2000.</ref>）。过去几十年在过去的几十年中，我们看到了一系列跨学科的算法的进步识别因果关系和效应大小的算法取得了一系列跨学科的进展。数据或混合实验和观察数据中识别因果关系和效应大小的算法取得了一系列跨学科的进展。这些发展有望使人们更好地利用适当的 "大数据"。它们已经被应用于基因组学、生态学、流行病学、空间物理学、临床医学、神经科学和许多其他领域，通常对其预测进行实验或准实验验证。在传统的因果关系研究中,当因果关系完全或部分已知时，用于识别因果效应或推断干预措施效果的算法，解决的是另一类问题。

=== 基于约束的因果发现方法 ===
基于约束的因果发现方法主要基于独立性和条件独立性检测判断变量之间的统计关系，并在一定的假设下先获得无向因果图。之后利用V-structure和一系列方向传播规则对因果图中的无向边进行方向传播。其经典方法包括PC<ref>Spirtes, P., Glymour, C. N., Scheines, R., & Heckerman, D. (2000). Causation, prediction, and search. MIT press.</ref>算法，能够处理隐变量（混淆变量）的FCI<ref>Spirtes, P., Meek, C., & Richardson, T. (1995, August). Causal inference in the presence of latent variables and selection bias. In Proceedings of the Eleventh conference on Uncertainty in artificial intelligence (pp. 499-506).</ref>算法，以及能够处理异构数据的CD-NOD<ref>Huang, B., Zhang, K., Zhang, J., Ramsey, J. D., Sanchez-Romero, R., Glymour, C., & Schölkopf, B. (2020). Causal Discovery from Heterogeneous/Nonstationary Data. J. Mach. Learn. Res., 21(89), 1-53.</ref>算法。

=== 基于评分的因果发现方法 ===
基于评分的方法通过对候选模型（即候选因果图）进行评分的方式，来选取评分最高的因果图作为结果。经典方法包括GES<ref>Chickering, D. M. (2002). Optimal structure identification with greedy search. Journal of machine learning research, 3(Nov), 507-554.</ref>以及一系列通过改变评分函数来适应不同任务的扩展方法，例如基于通用评分函数的因果发现方法<ref>Huang, B., Zhang, K., Lin, Y., Schölkopf, B., & Glymour, C. (2018, July). Generalized score functions for causal discovery. In Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (pp. 1551-1560).</ref>。此外，对于实际情况中Faithfulness假设不满足的情景，Exact Search<ref>Ng, Ignavier, Yujia Zheng, Jiji Zhang, and Kun Zhang. "Reliable Causal Discovery with Improved Exact Search and Weaker Assumptions." ''Advances in Neural Information Processing Systems'' 34 (2021): 20308-20320.</ref>可提供可靠的马尔科夫等价类。

=== 基于约束性函数因果模型的因果发现方法 ===
基于约束性函数因果模型的因果方法通过假设结果和原因之间的函数关系类别，从而判断其间的方向。经典方法包括线性非高斯无环模型<ref>Shimizu, S., Hoyer, P. O., Hyvärinen, A., Kerminen, A., & Jordan, M. (2006). A linear non-Gaussian acyclic model for causal discovery. Journal of Machine Learning Research, 7(10).</ref>，加性噪声模型<ref>Hoyer, P. O., Janzing, D., Mooij, J. M., Peters, J., & Schölkopf, B. (2008, December). Nonlinear causal discovery with additive noise models. In NIPS (Vol. 21, pp. 689-696).</ref>和后非线性模型<ref>Zhang, K., & Hyvärinen, A. (2009, June). On the Identifiability of the Post-Nonlinear Causal Model. In 25th Conference on Uncertainty in Artificial Intelligence (UAI 2009) (pp. 647-655). AUAI Press.</ref>。近期通过对混合函数进行限制从而得到非线性独立成分分析的可识别性的理论<ref>Zheng, Yujia, Ignavier Ng, and Kun Zhang. "On the Identifiability of Nonlinear ICA: Sparsity and Beyond." ''arXiv preprint arXiv:2206.07751'' (2022).</ref>也为通用的非线性的因果发现提供了新的方法。

=== 因果隐表征学习 ===
因果隐表征学习专注于隐变量与观测变量间的关系以及隐变量间的关系，经典工作包括GIN<ref>Xie, Cai, Huang, Glymour, Hao, Zhang, "Generalized Independent Noise Condition for Estimating Linear Non-Gaussian Latent Variable Causal Graphs," NeurIPS 2020</ref>以及层级化隐变量学习<ref>Feng Xie, Biwei Huang, Zhengming Chen, Yangbo He, Zhi Geng, Kun Zhang, "Estimation of Linear Non-Gaussian Latent Hierarchical Structure," accepted to ''International Conference on Machine Learning (ICML) 2022''</ref>。

=== 因果发现工具：causal-learn ===

Causal-learn<ref>https://causal-learn.readthedocs.io/en/latest/</ref>是Tetrad<ref>https://github.com/cmu-phil/tetrad</ref> java代码的一个Python翻译和扩展。它提供了最新的因果发现方法的实现，以及简单而直观的API。

因果发现

2022-06-26T06:03:14Z

Kunwuz：

Peter Spirtes

2022-06-26T05:44:35Z

Kunwuz：

=== 简介 ===
[[文件:Peter spirtes public photo.jpg|缩略图]]

Peter Spirtes是卡内基梅隆大学的Marianna Brown Dietrich教授和哲学系主任。他的研究兴趣跨越多个基础学科，涉及哲学、统计学、图论和计算机科学。他的研究对一些需要从统计数据中进行因果推断的学科有着深刻的影响。

Peter Spirtes与Clark Glymour一起提出了最早的因果发现算法之一，即PC。他还发表了该领域广泛使用的参考书之一（"Causation, Prediction, and Search<ref>Spirtes, Peter, et al. ''Causation, prediction, and search''. MIT press, 2000.</ref>"）。他的工作表明，在某些情况下，有一些计算机程序可以在合理的假设下从实验或非实验数据，或两者的组合中，可靠地得出有用的因果结论。他目前的研究集中如何将因果科学的基础假设弱化，从而将结果的应用扩展到更广泛，更通用的场景中。同时，他也关注因果发现在大规模数据上的引用。

Peter Spirtes的研究对包括生物学在内的许多不同学科都有重要的理论和实践意义。在理论上，它帮助我们理解了概率和因果关系之间的关系，以及在各种不同的假设下，从各种数据中进行可靠的因果推断的确切限度是什么。在实践上，它为科学家提供了一个有用的工具，帮助他们建立因果模型。

=== 代表工作 ===

Spirtes, P., Zhang, J. (forthcoming) “A Uniformly Consistent Estimator of Causal Effects Under The k-Triangle-Faithfulness Assumption”, ''Statistical Science''.

Spirtes, P., (2013) "Calculation of Entailed Rank Constraints in Partially Non-Linear and Cyclic Models", ''Proceedings of the Twenty-Ninth Conference Annual Conference on Uncertainty in Artificial Intelligence (UAI-13)'', AUAI Press, 2013, pp. 606-615.

Ramsey, J., Spirtes, P, and Glymour, C. (2011) “On meta-analyses of imaging data and the mixture of records.” ''NeuroImage'' '''57'''(2): 323-330.

Zhang, J., and Spirtes, P. (2011) "Intervention, Determinism, and the Causal Minimality Condition”, ''Synthese'', 2011, '''182:13''', pp. 335-347.

Ali, A., Richardson, T., Spirtes, P. (2009) “Markov Equivalence For Ancestral Graphs”, ''Annals of Statistics'', '''37''', 5B, 2808-2837.

Tillman, R., Gretton, A. and Spirtes, P. (2009) “Nonlinear directed acyclic structure learning with weakly additive noise models”, ''NIPS 2009''.

Spirtes, P. (2009) "Variable Definition and Causal Inference", ''Proceedings of the 13th International Congress of Logic Methodology and Philosophy of Science'', pp. 514-53.

Zhang, J., and Spirtes, P. (2009) "Detection of Unfaithfulness and Robust Causal Inference", ''Minds and Machines'', 18:2, pp. 239-272.

Silva, R., Glymour, C., Scheines, R. and Spirtes, P. (2006) “Learning the Structure of Latent Linear Structure Models,” ''Journal of Machine Learning Research'', 7, 191-246.

Ramsey, J., Zhang, J., and Spirtes, P., (2006) “Adjacency-Faithfulness and Conservative Causal Inference”, ''Uncertainty in Artificial Intelligence 2006'', Boston, MA.

Spirtes, P. (2005) “Graphical Models, Causal Inference, and Econometric Models”, ''Journal of Economic Methodology''. 2005 '''12:1''', pp. 1–33.

Zhang, J., and Spirtes, P. (2005) “A Transformational Characterization of Markov Equivalence for Directed Acyclic Graphs with Latent Variables”, ''Uncertainty in Artificial Intelligence 2005'', Edinboro, Scotland.

Ali, R., Richardson, T., Spirtes, P., and Zhang, J. (2005) “Towards Characterizing Markov Equivalence Classes for Directed Acyclic Graph Models with Latent Variables”, ''Uncertainty in Artificial Intelligence 2005'', Edinboro, Scotland.

Spirtes, P., and Scheines, R. (2004). “Causal Inference of Ambiguous Manipulations”, in ''Proceedings of the Philosophy of Science Association Meetings'', 2002.

Chu, T., Glymour, C., Scheines, R., Spirtes, P. (2003) “A Statistical Problem for Inference to Regulatory Structure from Associations of Gene Expression Measurements with Microarrays”, ''Bioinformatics'', '''19''', pp. 1147-1152.

Robins, J., Scheines, R., Spirtes, P., and Wasserman, L. (2003). “Uniform Consistency in Causal Inference”, ''Biometrika'', September, '''90''': pp. 491 – 515.

Zhang, J., and Spirtes, P. (2003) “Strong Faithfulness and Uniform Consistency in Causal Inference”, ''UAI '03, Proceedings of the 19th Conference in Uncertainty in Artificial Intelligence'', August 7-10 2003, Acapulco, Mexico, ed. by Christopher Meek and Uffe Kjarulff, Morgan Kaufmann.

Richardson, T., Spirtes, P. (2002) “Ancestral Graph Markov Models”, ''Annals of Statistics'', 2002, '''30''' pp. 962-1030.

Spirtes, P., Glymour, C. and Scheines, R. (2000). Causation, Prediction, and Search, 2nd ed. New York, N.Y.: MIT Press.

Spirtes, P., Glymour, C., and Scheines, R. (2000) Constructing Bayesian Network Models of Gene Expression Networks from Microarray Data, to appear in the Proceedings of the Atlantic Symposium on Computational Biology, Genome Information Systems & Technology.

Robins, J., Scheines, R., Spirtes, P., and Wasserman, L. (2000) Uniform Consistency in Causal Inference, Carnegie Mellon University Department of Statistics Technical Report 725.

Richardson, T., and Spirtes, P. (2000) Ancestral Markov Graphical Models, University of Washington Department of Statistics Technical Report 375.

Spirtes, P. (2000) An Anytime Algorithm for Causal Inference, to be presented at AI and Statistics 2001.

Spirtes, P. (1997). Limits on Causal Inference from Statistical Data, presented at American Economics Association Meeting.

Spirtes, P., Cooper, G. (1997). An Experiment in Causal Discovery Using a Pneumonia Database, Proceedings of AI and Statistics 99.

Spirtes, P., Richardson, T., Meek, C. (1997). The Dimensionality of Mixed Ancestral Graphs, Technical Report CMU-83-Phil.

Spirtes, P., Richardson, T., Meek, C., Scheines, R., and Glymour, C. (1997). Using Path Diagrams as a Structural Equation Modelling Tool, Technical Report CMU-82-Phil.

Scheines, R., Spirtes, P., Glymour, C., Meek, C., and Richardson, T. (forthcoming). The TETRAD Project: Constraint Based Aids to Causal Model Specification, Multivariate Behavioral Research

Spirtes, P., Glymour, C. and Scheines, R. (1993). Causation, Prediction, and Search, New York, N.Y.: Springer-Verlag.

Scheines, R. (forthcoming). An Introduction to Causal Inference, in Causality in Crisis, ed. by Steven Turner and Vaughan McKim, University of Notre Dame Press.

Spirtes, P., Richardson, T., Meek, C., Scheines, R., and Glymour, C., (1996). Using D-separation to Calculate Zero Partial Correlations in Linear Models with Correlated Errors, Technical Report CMU-72-Phil.

Spirtes, P., and Richardson, T. (1996). A Polynomial Time Algorithm For Determining DAG Equivalence in the Presence of Latent Variables and Selection Bias, Proceedings of the 6th International Workshop on Artificial Intelligence and Statistics.

Spirtes, P., Richardson, T., and Meek, C. (1996). Heuristic Greedy Search Algorithms for Latent Variable Models, Proceedings of the 6th International Workshop on Artificial Intelligence and Statistics.

Richardson, T., and Spirtes, P. (1996). Automated discovery of linear feedback models, Technical Report CMU-75-Phil.

Spirtes, P., and Scheines, R. (forthcoming). Reply to Freedman, in Causality in Crisis, ed. by Steven Turner and Vaughan McKim, University of Notre Dame Press.

Spirtes, P., Meek, C., and Richardson, T. (1996). Causal Inference in the Presence of Latent Variables and Selection Bias, Technical Report CMU-77-Phil.

Spirtes, P. (1995). Directed Cyclic Graphical Representation of Feedback Models, Proceedings of the Eleventh Conference on Uncertainty in Artificial Intelligence, ed. by Philippe Besnard and Steve Hanks, Morgan Kaufmann Publishers, Inc., San Mateo, 1995.

Peter Spirtes

2022-06-18T13:14:11Z

Kunwuz：

Peter Spirtes

2022-06-18T13:11:55Z

Kunwuz：/* 简介 */

Peter Spirtes

2022-06-18T10:10:34Z

Kunwuz：

文件:Peter spirtes public photo.jpg

2022-06-18T10:10:14Z

Kunwuz：

Peter Spirtes's photo from his website.