MMD-B-Fair: Learning Fair Representations with Statistical Testing

TL;DR

MMD-B-Fair introduces a kernel two-sample testing approach to learn fair data representations that conceal sensitive information while retaining target attribute information, avoiding complex adversarial methods.

Contribution

It presents a novel, efficient method for fair representation learning based on statistical testing, bypassing adversarial and generative models used previously.

Findings

Successfully hides sensitive attribute information in representations.

Maintains target attribute information for downstream tasks.

Outperforms existing methods in fairness and utility metrics.

Abstract

We introduce a method, MMD-B-Fair, to learn fair representations of data via kernel two-sample testing. We find neural features of our data where a maximum mean discrepancy (MMD) test cannot distinguish between representations of different sensitive groups, while preserving information about the target attributes. Minimizing the power of an MMD test is more difficult than maximizing it (as done in previous work), because the test threshold's complex behavior cannot be simply ignored. Our method exploits the simple asymptotics of block testing schemes to efficiently find fair representations without requiring complex adversarial optimization or generative modelling schemes widely used by existing work on fair representation learning. We evaluate our approach on various datasets, showing its ability to ``hide'' information about sensitive attributes, and its effectiveness in downstream transfer tasks.