Probabilistic Bias Mitigation in Word Embeddings

By Hailey James, David Alvarez-Melis. 2019

4 citations recorded by the source corpus.

It has been shown that word embeddings derived from large corpora tend to incorporate biases present in their training data. Various methods for mitigating these biases have been proposed, but recent work has demonstrated that these methods hide but fail to truly remove the biases, which can still be observed in word nearest-neighbor statistics. In this work we propose a probabilistic view of word embedding bias. We leverage this framework to present a novel method for mitigating bias which relies on probabilistic observations to yield a more robust bias mitigation algorithm. We demonstrate that this method effectively reduces bias according to three separate measures of bias while maintaining embedding quality across various popular benchmark semantic tasks

Subfields: Representation & Self-Supervised Learning, Safety, Robustness & Interpretability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Representation & Self-Supervised Learning

  1. Hal Daumé Iii
  2. Hanna M. Wallach
  3. Kate Saenko
  4. Qiang Yang
  5. Mohammad Rostami
  6. Trevor Darrell
  7. Ramalingam Chellappa
  8. Gabriella Csurka

In Safety, Robustness & Interpretability

  1. Su Lin Blodgett
  2. Hal Daumé Iii
  3. Hanna M. Wallach
  4. Solon Barocas
  5. Emilio Ferrara
  6. Yulia Tsvetkov
  7. Yejin Choi
  8. Sunipa Dev

Citations in the corpus