Abstract
With a large amount of data accumulated in many fields, causal discovery based on observational data is gradually emerging, which is considered to be the basis for realizing strong artificial intelligence. However, the existing main causal discovery methods, including constraint-based methods, structural causal model based methods, and scoring-based methods, cannot find real causal relations accurately and quickly. In this paper, we propose a causal discovery method based on genetic algorithm, which combines structural causal model, scoring method, and genetic search algorithm. The core of our method is to divide the causal relation discovery process into the evaluation phase based on the features of structural causal model and the search phase based on the genetic algorithm. In the evaluation phase, the causal graph is evaluated from three aspects: model deviation, noise independence, and causal graph cyclicity, which effectively ensures the accuracy of causal discovery. In the search phase, an efficient random search is designed based on genetic algorithm, which greatly improves the causal discovery efficiency. This paper implements the corresponding algorithm, namely SCM-GA (Structural Causal Model based Genetic Algorithm), and conducts experiments on several simulated datasets and one widely used real-scene dataset. The experiments compare five classic baseline algorithms, and the results show that SCM-GA has achieved great improvement in accuracy, applicability, and efficiency. Especially on the real scene dataset, SCM-GA achieves better results than the state-of-the-art algorithm, with similar SHD (Structure Hamming Distance) value, 40% higher recall rate, and 83.3% shorter running time.
Supported by National Science and Technology Major Project (Grant No. 2020AAA0109401).
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
It is commonly referred to as causal discovery, this abbreviation is also used in related literature, so in this article we do the same.
References
Cai, R.C., Chen, W., Zhang, K., Hao, Z.F.: A survey on non-temporal series observational data based causal discovery (in Chinese). Chin. J. Comput. 40(6), 1470–1490 (2017)
Cai, R.C., Hao, Z.F.: Casual discovery in big data (in Chinese). Science Press (2018)
Chickering, D.M.: Learning Bayesian networks is NP-complete. In: Fisher, D., Lenz, H.J. (eds.) Learning from Data. Lecture Notes in Statistics, vol. 112, pp. 121–130. Springer, New York, NY (1996). https://doi.org/10.1007/978-1-4612-2404-4_12
Chickering, D.M.: Optimal structure identification with greedy search. J. Mach. Learn. Res. 3(3), 507–554 (2002). https://doi.org/10.1162/153244303321897717
Chickering, M., Heckerman, D., Meek, C.: Large-sample learning of Bayesian networks is NP-hard. J. Mach. Learn. Res. 5, 1287–1330 (2004)
Glymour, C., Zhang, K., Spirtes, P.: Review of causal discovery methods based on graphical models. Front. Genet. 10, 524 (2019). https://doi.org/10.3389/fgene.2019.00524
Goudet, O., Kalainathan, D., Caillou, P., Guyon, I., Lopez-Paz, D., Sebag, M.: Learning functional causal models with generative neural networks. In: Escalante, H.J., et al. (eds.) Explainable and Interpretable Models in Computer Vision and Machine Learning. TSSCML, pp. 39–80. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-98131-4_3
Gretton, A., Herbrich, R., Smola, A., Bousquet, O., Schölkopf, B., et al.: Kernel methods for measuring independence. J. Mach. Learn. Res. 6, 2075–2129 (2005). https://doi.org/10.1007/s10846-005-9001-9
Haughton, D.M.: On the choice of a model to fit data from an exponential family. Ann. Stat. 16, pp. 342–355 (1988). https://doi.org/10.1214/aos/1176350709
He, Y.-B., Geng, Z., Liang, X.: Learning causal structures based on Markov equivalence class. In: Jain, S., Simon, H.U., Tomita, E. (eds.) ALT 2005. LNCS (LNAI), vol. 3734, pp. 92–106. Springer, Heidelberg (2005). https://doi.org/10.1007/11564089_9
He, Y., Cui, P., Shen, Z., Xu, R., Liu, F., Jiang, Y.: Daring: differentiable causal discovery with residual independence. In: Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery Data Mining, pp. 596–605 (2021). https://doi.org/10.1145/3447548.3467439
Hoyer, P., Janzing, D., Mooij, J.M., Peters, J., Schölkopf, B.: Nonlinear causal discovery with additive noise models. In: Advances in Neural Information Processing Systems 21 (2008)
Kitson, N.K., Constantinou, A.C., Guo, Z., Liu, Y., Chobtham, K.: A survey of Bayesian network structure learning. arXiv preprint arXiv:2109.11415 (2021)
Lachapelle, S., Brouillard, P., Deleu, T., Lacoste-Julien, S.: Gradient-based neural DAG learning. In: International Conference on Learning Representations (2020)
Mattmann, C.A.: A vision for data science. Nature 493(7433), 473–475 (2013). https://doi.org/10.1038/493473a
Peters, J., Janzing, D., Schölkopf, B.: Elements of causal inference: foundations and learning algorithms. The MIT Press (2017)
Peters, J., Mooij, J.M., Janzing, D., Schölkopf, B.: Causal discovery with continuous additive noise models. J. Mach. Learn. Res. 15(58), 2009–2053 (2014). http://hdl.handle.net/2066/130001
Sachs, K., Perez, O., Pe’er, D., Lauffenburger, D.A., Nolan, G.P.: Causal protein-signaling networks derived from multiparameter single-cell data. Science 308(5721), 523–529 (2005). https://doi.org/10.1126/science.1105809
Shimizu, S., Hoyer, P.O., Hyvärinen, A., Kerminen, A., Jordan, M.: A linear non-gaussian acyclic model for causal discovery. J. Mach. Learn. Res. 7(10), 2003–2030 (2006). https://doi.org/10.1007/s10883-006-0005-y
Spirtes, P., Glymour, C., Scheines, R.: Causality from probability (1989)
Spirtes, P., Glymour, C.N., Scheines, R., Heckerman, D.: Causation, prediction, and search. MIT press (2000)
Verma, T.S., Pearl, J.: Equivalence and synthesis of causal models. In: Probabilistic and Causal Inference: the works of Judea Pearl, pp. 221–236 (2022). https://doi.org/10.1145/3501714.3501732
Wang, X., Dunson, D., Leng, C.: No penalty no tears: least squares in high-dimensional linear models. In: International Conference on Machine Learning, pp. 1814–1822. PMLR (2016)
Zhang, K., Hyvärinen, A.: Causality discovery with additive disturbances: an information-theoretical perspective. In: Buntine, W., Grobelnik, M., Mladenić, D., Shawe-Taylor, J. (eds.) ECML PKDD 2009. LNCS (LNAI), vol. 5782, pp. 570–585. Springer, Heidelberg (2009). https://doi.org/10.1007/978-3-642-04174-7_37
Zhang, K., Hyvärinen, A.: Distinguishing causes from effects using nonlinear acyclic causal models. In: Causality: Objectives and Assessment, pp. 157–164. PMLR (2010)
Zhang, K., Hyvarinen, A.: On the identifiability of the post-nonlinear causal model. arXiv preprint arXiv:1205.2599 (2012)
Zheng, X., Aragam, B., Ravikumar, P.K., Xing, E.P.: DAGs with no tears: continuous optimization for structure learning. In: Advances in Neural Information Processing Systems 31 (2018)
Zhou, S.: Thresholding procedures for high dimensional variable selection and statistical estimation. In: Advances in Neural Information Processing Systems 22 (2009)
Zhu, S., Ng, I., Chen, Z.: Causal discovery with reinforcement learning. In: International Conference on Learning Representations (2020)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Chen, Z., Liu, K., Jiao, W. (2022). A Genetic Algorithm for Causal Discovery Based on Structural Causal Model. In: Fang, L., Povey, D., Zhai, G., Mei, T., Wang, R. (eds) Artificial Intelligence. CICAI 2022. Lecture Notes in Computer Science(), vol 13606. Springer, Cham. https://doi.org/10.1007/978-3-031-20503-3_4
Download citation
DOI: https://doi.org/10.1007/978-3-031-20503-3_4
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-20502-6
Online ISBN: 978-3-031-20503-3
eBook Packages: Computer ScienceComputer Science (R0)