Fair Clustering via Hierarchical Fair-Dirichlet Process

TL;DR

This paper introduces a novel model-based approach to fair clustering, addressing the need for balanced representation of protected attributes within clusters, expanding beyond existing objective-based methods.

Contribution

It proposes a new model-based formulation for fair clustering, complementing the prevalent objective optimization approaches in the literature.

Findings

Develops a hierarchical fair-Dirichlet process model

Demonstrates improved fairness in clustering results

Provides a theoretical framework for fair clustering

Abstract

The advent of ML-driven decision-making and policy formation has led to an increasing focus on algorithmic fairness. As clustering is one of the most commonly used unsupervised machine learning approaches, there has naturally been a proliferation of literature on {\em fair clustering}. A popular notion of fairness in clustering mandates the clusters to be {\em balanced}, i.e., each level of a protected attribute must be approximately equally represented in each cluster. Building upon the original framework, this literature has rapidly expanded in various aspects. In this article, we offer a novel model-based formulation of fair clustering, complementing the existing literature which is almost exclusively based on optimizing appropriate objective functions.