English

A Scalable Algorithm for Individually Fair K-means Clustering

Data Structures and Algorithms 2024-02-14 v1 Computers and Society Machine Learning

Abstract

We present a scalable algorithm for the individually fair (pp, kk)-clustering problem introduced by Jung et al. and Mahabadi et al. Given nn points PP in a metric space, let δ(x)\delta(x) for xPx\in P be the radius of the smallest ball around xx containing at least n/kn / k points. A clustering is then called individually fair if it has centers within distance δ(x)\delta(x) of xx for each xPx\in P. While good approximation algorithms are known for this problem no efficient practical algorithms with good theoretical guarantees have been presented. We design the first fast local-search algorithm that runs in ~O(nk2)O(nk^2) time and obtains a bicriteria (O(1),6)(O(1), 6) approximation. Then we show empirically that not only is our algorithm much faster than prior work, but it also produces lower-cost solutions.

Keywords

Cite

@article{arxiv.2402.06730,
  title  = {A Scalable Algorithm for Individually Fair K-means Clustering},
  author = {MohammadHossein Bateni and Vincent Cohen-Addad and Alessandro Epasto and Silvio Lattanzi},
  journal= {arXiv preprint arXiv:2402.06730},
  year   = {2024}
}

Comments

32 pages, 2 figures, to appear at the 27th International Conference on Artificial Intelligence and Statistics (AISTATS) 2024

R2 v1 2026-06-28T14:44:33.692Z