Applying dependency patterns in causal discovery of latent variable models

Xuhui Zhang, Kevin B. Korb, Ann E. Nicholson, Steven Mascaro

    Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

    1 Citation (Scopus)

    Abstract

    Latent variables represent unmeasured causal factors. Some, such as intelligence, cannot be directly measured; others may be, but we do not know about them or know how to measure them when making our observations. Regardless, in many cases, the influence of latent variables is real and important, and optimal modeling cannot be done without them. However, in many of those cases the influence of latent variables reveals itself in patterns of measured dependency that cannot be repro- duced using the observed variables alone, under the assumptions of the causal Markov property and faithfulness. In such cases, latent variables may be posited to the advantage of the causal discovery process. All latent variable discovery takes advantage of this; we make the process explicit.

    Original languageEnglish
    Title of host publicationArtificial Life and Computational Intelligence
    Subtitle of host publicationThird Australasian Conference, ACALCI 2017, Geelong, VIC, Australia, January 31 – February 2, 2017, Proceedings
    EditorsMarkus Wagner, Xiaodong Li, Tim Hendtlass
    Place of PublicationCham, Switzerland
    PublisherSpringer
    Pages134-143
    Number of pages10
    ISBN (Electronic)9783319516912
    ISBN (Print)9783319516905
    DOIs
    Publication statusPublished - 2017
    EventAustralasian Conference on Artificial Life and Computational Intelligence (ACALCI) 2017 - Deakin University, Waterfront Campus, Geelong, Australia
    Duration: 31 Jan 20172 Feb 2017
    Conference number: 3rd
    http://www.acalci.net/2017/
    https://link.springer.com/book/10.1007/978-3-319-51691-2 (Springer Proceedings)

    Publication series

    NameLecture Notes in Artificial Intelligence
    PublisherSpringer
    Volume10142
    ISSN (Print)0302-9743
    ISSN (Electronic)1611-3349

    Conference

    ConferenceAustralasian Conference on Artificial Life and Computational Intelligence (ACALCI) 2017
    Abbreviated titleACALCI 2017
    CountryAustralia
    CityGeelong
    Period31/01/172/02/17
    OtherACACLI 2017 is co-located with the Australasian Computer Science Week (ACSW 2017), which will be held at Deakin University's Waterfront Campus, Geelong, which is about 70 kilometers west of Mebourne.

    3rd Australasian Conference on Artificial Life and Computational Intelligence, ACALCI 2017
    Internet address

    Keywords

    • Bayesian networks
    • Causal discovery
    • Latent variables

    Cite this

    Zhang, X., Korb, K. B., Nicholson, A. E., & Mascaro, S. (2017). Applying dependency patterns in causal discovery of latent variable models. In M. Wagner, X. Li, & T. Hendtlass (Eds.), Artificial Life and Computational Intelligence: Third Australasian Conference, ACALCI 2017, Geelong, VIC, Australia, January 31 – February 2, 2017, Proceedings (pp. 134-143). (Lecture Notes in Artificial Intelligence; Vol. 10142). Springer. https://doi.org/10.1007/978-3-319-51691-2_12