Fair Hierarchical Clustering

TL;DR

This paper introduces efficient algorithms for fair hierarchical clustering that ensure equitable data partitioning while maintaining high clustering quality, addressing fairness concerns in recursive data segmentation.

Contribution

It extends fairness notions to hierarchical clustering and provides provably good, efficient algorithms for achieving fair data partitions.

Findings

Algorithms find fair hierarchical clusterings with negligible loss in objective

Proven theoretical guarantees for algorithm performance

Empirical results demonstrate effectiveness in real datasets

Abstract

As machine learning has become more prevalent, researchers have begun to recognize the necessity of ensuring machine learning systems are fair. Recently, there has been an interest in defining a notion of fairness that mitigates over-representation in traditional clustering. In this paper we extend this notion to hierarchical clustering, where the goal is to recursively partition the data to optimize a specific objective. For various natural objectives, we obtain simple, efficient algorithms to find a provably good fair hierarchical clustering. Empirically, we show that our algorithms can find a fair hierarchical clustering, with only a negligible loss in the objective.