Identifying and reducing gender bias in word-level language models

Shikha Bordia, Samuel R. Bowman

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Abstract

    Many text corpora exhibit socially problematic biases, which can be propagated or amplified in the models trained on such data. For example, doctor cooccurs more frequently with male pronouns than female pronouns. In this study we (i) propose a metric to measure gender bias; (ii) measure bias in a text corpus and the text generated from a recurrent neural network language model trained on the text corpus; (iii) propose a regularization loss term for the language model that minimizes the projection of encoder-trained embeddings onto an embedding subspace that encodes gender; (iv) finally, evaluate efficacy of our proposed method on reducing gender bias. We find this regularization method to be effective in reducing gender bias up to an optimal weight assigned to the loss term, beyond which the model becomes unstable as the perplexity increases. We replicate this study on three training corpora-Penn Treebank,WikiText-2, and CNN/Daily Mail-resulting in similar conclusions.

    Original languageEnglish (US)
    Title of host publicationNAACL HLT 2019 - 2019 Conference of the North American Chapter of the Association for Computational Linguistics
    Subtitle of host publicationHuman Language Technologies - Proceedings of the Student Research Workshop
    PublisherAssociation for Computational Linguistics (ACL)
    Pages7-15
    Number of pages9
    ISBN (Electronic)9781950737154
    StatePublished - 2019
    Event2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL HLT 2019 - Student Research Workshop, SRW 2019 - Minneapolis, United States
    Duration: Jun 3 2019Jun 5 2019

    Publication series

    NameNAACL HLT 2019 - 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies - Proceedings of the Student Research Workshop

    Conference

    Conference2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL HLT 2019 - Student Research Workshop, SRW 2019
    CountryUnited States
    CityMinneapolis
    Period6/3/196/5/19

    ASJC Scopus subject areas

    • Language and Linguistics
    • Computer Science Applications
    • Linguistics and Language

    Fingerprint Dive into the research topics of 'Identifying and reducing gender bias in word-level language models'. Together they form a unique fingerprint.

  • Cite this

    Bordia, S., & Bowman, S. R. (2019). Identifying and reducing gender bias in word-level language models. In NAACL HLT 2019 - 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies - Proceedings of the Student Research Workshop (pp. 7-15). (NAACL HLT 2019 - 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies - Proceedings of the Student Research Workshop). Association for Computational Linguistics (ACL).