Learning Smooth and Fair Representations

Xavier Gitiaux · Huzefa Rangwala

Keywords: [ Ethics and Safety ] [ Fairness, Equity, Justice, and Safety ]

[ Abstract ]
Wed 14 Apr 12:45 p.m. PDT — 2:45 p.m. PDT
Oral presentation: Fairness / Privacy / Decision Making / Data Cleaning
Thu 15 Apr noon PDT — 1 p.m. PDT


This paper explores the statistical properties of fair representation learning, a pre-processing method that preemptively removes the correlations between features and sensitive attributes by mapping features to a fair representation space. We show that the demographic parity of a representation can be certified from a finite sample if and only if the mapping guarantees that the chi-squared mutual information between features and representations is finite for distributions of the features. Empirically, we find that smoothing representations with an additive Gaussian white noise provides generalization guarantees of fairness certificates, which improves upon existing fair representation learning approaches.

Chat is not available.