English

Breaching the 2 LMP Approximation Barrier for Facility Location with Applications to k-Median

Data Structures and Algorithms 2026-01-29 v2

Abstract

The Uncapacitated Facility Location (UFL) problem is one of the most fundamental clustering problems: Given a set of clients CC and a set of facilities FF in a metric space (CF,dist)(C \cup F, dist) with facility costs open:FR+open : F \to \mathbb{R}^+, the goal is to find a set of facilities SFS \subseteq F to minimize the sum of the opening cost open(S)open(S) and the connection cost d(S):=pCmincSdist(p,c)d(S) := \sum_{p \in C} \min_{c \in S} dist(p, c). An algorithm for UFL is called a Lagrangian Multiplier Preserving (LMP) α\alpha approximation if it outputs a solution SFS\subseteq F satisfying open(S)+d(S)open(S)+αd(S)open(S) + d(S) \leq open(S^*) + \alpha d(S^*) for any SFS^* \subseteq F. The best-known LMP approximation ratio for UFL is at most 22 by the JMS algorithm of Jain, Mahdian, and Saberi based on the Dual-Fitting technique. We present a (slightly) improved LMP approximation algorithm for UFL. This is achieved by combining the Dual-Fitting technique with Local Search, another popular technique to address clustering problems. From a conceptual viewpoint, our result gives a theoretical evidence that local search can be enhanced so as to avoid bad local optima by choosing the initial feasible solution with LP-based techniques. Using the framework of bipoint solutions, our result directly implies a (slightly) improved approximation for the kk-Median problem from 2.6742 to 2.67059.

Keywords

Cite

@article{arxiv.2207.05150,
  title  = {Breaching the 2 LMP Approximation Barrier for Facility Location with Applications to k-Median},
  author = {Vincent Cohen-Addad and Fabrizio Grandoni and Euiwoong Lee and Chris Schwiegelshohn},
  journal= {arXiv preprint arXiv:2207.05150},
  year   = {2026}
}

Comments

Conference version in SODA2023. This version corrects a confusing bug in the definition of the upperbounding LP