A Causal Inference Method for Reducing Gender Bias in Word Embedding Relations

Zekun Yang, Juan Feng*

*Corresponding author for this work

Research output: Chapters, Conference Papers, Creative and Literary WorksRGC 32 - Refereed conference paper (with host publication)peer-review

Abstract

Word embedding has become essential for natural language processing as it boosts empirical performances of various tasks. However, recent research discovers that gender bias is incorporated in neural word embeddings, and downstream tasks that rely on these biased word vectors also produce gender-biased results. While some word-embedding gender-debiasing methods have been developed, these methods mainly focus on reducing gender bias associated with gender direction and fail to reduce the gender bias presented in word embedding relations. In this paper, we design a causal and simple approach for mitigating gender bias in word vector relation by utilizing the statistical dependency between gender-definition word embeddings and gender-biased word embeddings. Our method attains state-of-the-art results on gender-debiasing tasks, lexical- and sentence-level evaluation tasks, and downstream coreference resolution tasks.
Original languageEnglish
Title of host publicationThe Thirty-Fourth AAAI Conference on Artificial Intelligence (AAAI-20)
Place of PublicationCalifornia
PublisherAAAI Press
Pages9434-9441
ISBN (Print)9781577358350 (10 issue set)
DOIs
Publication statusPublished - Feb 2020
Event34th AAAI Conference on Artificial Intelligence (AAAI-20) - New York, United States
Duration: 7 Feb 202012 Feb 2020
https://aaai.org/Conferences/AAAI-20/
https://aaai.org/ojs/index.php/AAAI/index

Publication series

NameProceedings of the AAAI Conference on Artificial Intelligence
PublisherAAAI Press
Number5
Volume34
ISSN (Print)2159-5399
ISSN (Electronic)2374-3468

Conference

Conference34th AAAI Conference on Artificial Intelligence (AAAI-20)
PlaceUnited States
CityNew York
Period7/02/2012/02/20
Internet address

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

  1. SDG 5 - Gender Equality
    SDG 5 Gender Equality

Fingerprint

Dive into the research topics of 'A Causal Inference Method for Reducing Gender Bias in Word Embedding Relations'. Together they form a unique fingerprint.

Cite this