Score: 0

Matrix Editing Meets Fair Clustering: Parameterized Algorithms and Complexity

Published: December 3, 2025 | arXiv ID: 2512.03718v1

By: Robert Ganian, Hung P. Hoang, Simon Wietheger

Potential Business Impact:

Makes computer grouping fair, even when hard.

Business Areas:
A/B Testing Data and Analytics

We study the computational problem of computing a fair means clustering of discrete vectors, which admits an equivalent formulation as editing a colored matrix into one with few distinct color-balanced rows by changing at most $k$ values. While NP-hard in both the fairness-oblivious and the fair settings, the problem is well-known to admit a fixed-parameter algorithm in the former ``vanilla'' setting. As our first contribution, we exclude an analogous algorithm even for highly restricted fair means clustering instances. We then proceed to obtain a full complexity landscape of the problem, and establish tractability results which capture three means of circumventing our obtained lower bound: placing additional constraints on the problem instances, fixed-parameter approximation, or using an alternative parameterization targeting tree-like matrices.

Country of Origin
🇦🇹 Austria

Page Count
15 pages

Category
Computer Science:
Data Structures and Algorithms