Parameterized Complexity of Fair Many-to-One Matchings

TL;DR

This paper studies the computational complexity of fair many-to-one matchings in bipartite graphs, showing which structural parameters make the problem tractable or hard, with implications for algorithm design.

Contribution

It provides a comprehensive parameterized complexity analysis, identifying both hardness results and fixed-parameter tractability for various graph parameters.

Findings

W[1]-hard with respect to feedback vertex number, tree-depth, max degree of U, combined

W[1]-hard with respect to path-width, number of groups, max degree of U, combined

FPT with respect to treewidth and max degree of V, neighborhood diversity, tree-depth and number of groups

Abstract

Given a bipartite graph $G=(U\cup V,E)$, a left-perfect many-to-one matching is a subset $M \subseteq E$ such that each vertex in $U$ is incident with exactly one edge in $M$. If $U$ is partitioned into some groups, the matching is called fair if for every $v\in V$, the difference between the number of vertices matched with $v$ in any two groups does not exceed a given threshold. In this paper, we investigate parameterized complexity of fair left-perfect many-to-one matching problem with respect to the structural parameters of the input graph. In particular, we prove that the problem is W[1]-hard with respect to the feedback vertex number, tree-depth and the maximum degree of $U$, combined. Also, it is W[1]-hard with respect to the path-width, the number of groups and the maximum degree of $U$, combined. In the positive side, we prove that the problem is FPT with respect to the treewidth and the maximum degree of $V$. Also, it is FPT with respect to the neighborhood diversity of the input graph (which implies being FPT with respect to vertex cover and modular-width). Finally, we prove that the problem is FPT with respect to the tree-depth and the number of groups.