Quantifying and Reducing Stereotypes in Word Embeddings

TL;DR

This paper investigates gender stereotypes in word embeddings, quantifies their extent using a novel analogy task, and proposes an efficient method to reduce such biases while maintaining embedding quality.

Contribution

It introduces a new framework for measuring gender bias in word embeddings and presents an algorithm to effectively reduce stereotypes with minimal training data.

Findings

Embeddings contain significant gender stereotypes, especially in professions.

The proposed algorithm reduces stereotypes while preserving embedding properties.

The framework can be adapted to other types of biases.

Abstract

Machine learning algorithms are optimized to model statistical properties of the training data. If the input data reflects stereotypes and biases of the broader society, then the output of the learning algorithm also captures these stereotypes. In this paper, we initiate the study of gender stereotypes in {\em word embedding}, a popular framework to represent text data. As their use becomes increasingly common, applications can inadvertently amplify unwanted stereotypes. We show across multiple datasets that the embeddings contain significant gender stereotypes, especially with regard to professions. We created a novel gender analogy task and combined it with crowdsourcing to systematically quantify the gender bias in a given embedding. We developed an efficient algorithm that reduces gender stereotype using just a handful of training examples while preserving the useful geometric properties of the embedding. We evaluated our algorithm on several metrics. While we focus on male/female stereotypes, our framework may be applicable to other types of embedding biases.