English

Fitting Tree Metrics with Minimum Disagreements

Data Structures and Algorithms 2023-08-01 v1

Abstract

In the L0L_0 Fitting Tree Metrics problem, we are given all pairwise distances among the elements of a set VV and our output is a tree metric on VV. The goal is to minimize the number of pairwise distance disagreements between the input and the output. We provide an O(1)O(1) approximation for L0L_0 Fitting Tree Metrics, which is asymptotically optimal as the problem is APX-Hard. For p1p\ge 1, solutions to the related LpL_p Fitting Tree Metrics have typically used a reduction to LpL_p Fitting Constrained Ultrametrics. Even though in FOCS '22 Cohen-Addad et al. solved L0L_0 Fitting (unconstrained) Ultrametrics within a constant approximation factor, their results did not extend to tree metrics. We identify two possible reasons, and provide simple techniques to circumvent them. Our framework does not modify the algorithm from Cohen-Addad et al. It rather extends any ρ\rho approximation for L0L_0 Fitting Ultrametrics to a 6ρ6\rho approximation for L0L_0 Fitting Tree Metrics in a blackbox fashion.

Keywords

Cite

@article{arxiv.2307.16066,
  title  = {Fitting Tree Metrics with Minimum Disagreements},
  author = {Evangelos Kipouridis},
  journal= {arXiv preprint arXiv:2307.16066},
  year   = {2023}
}

Comments

Accepted at ESA Track S 2023

R2 v1 2026-06-28T11:43:33.950Z