UniFair: A unified fair clustering approach based on separation and compactness
2026-06-03 • Machine Learning
Machine Learning
AI summaryⓘ
The authors address fairness in clustering, where groups might be treated unequally. They introduce UniFair, a method that balances two types of fairness: making sure groups are clearly separated from decision boundaries (separation fairness) and ensuring groups have similar clustering quality (social fairness). Their approach uses gradient-based optimization and can work with traditional and deep learning clustering. Tests on various datasets show UniFair reduces fairness issues without much loss in clustering performance.
clusteringk-meansfairnessseparation fairnesssocial fairnessdecision boundariesgradient-based optimizationdeep clusteringautoencoderwithin-cluster distortion
Authors
Antonia Karra, Vasiliki Papanikou, Georgios Vardakas, Evaggelia Pitoura, Aristidis Likas
Abstract
Clustering is increasingly used to support high-impact decisions, yet standard objectives such as $k$-means can produce clusterings that treat demographic groups unequally. Existing fair clustering methods typically optimize a single notion of fairness and often overlook how clustering costs interact with the geometry of the induced decision boundaries. We propose \textsc{UniFair}, a unified framework that jointly optimizes \emph{separation fairness} and \emph{social fairness}. Separation fairness encourages protected groups to lie farther from the induced decision boundaries, while social fairness reduces disparities in within-cluster distortion by penalizing group-wise clustering costs. We develop gradient-based optimization procedures for separation-fair and unified $k$-means objectives, and extend them to deep clustering by enforcing the same criteria in the latent space of an autoencoder. Experiments on tabular and image datasets show that \textsc{UniFair} reduces both boundary-related and cost-based group disparities with only a modest increase in clustering loss.