English

Deterministic metric $1$-median selection with very few queries

Data Structures and Algorithms 2022-01-06 v1 Computational Complexity Discrete Mathematics

Abstract

Given an nn-point metric space (M,d)(M,d), {\sc metric 11-median} asks for a point pMp\in M minimizing xMd(p,x)\sum_{x\in M}\,d(p,x). We show that for each computable function f ⁣:Z+Z+f\colon \mathbb{Z}^+\to\mathbb{Z}^+ satisfying f(n)=ω(1)f(n)=\omega(1), {\sc metric 11-median} has a deterministic, o(n)o(n)-query, o(f(n)logn)o(f(n)\cdot\log n)-approximation and nonadaptive algorithm. Previously, no deterministic o(n)o(n)-query o(n)o(n)-approximation algorithms are known for {\sc metric 11-median}. On the negative side, we prove each deterministic O(n)O(n)-query algorithm for {\sc metric 11-median} to be not (δlogn)(\delta\log n)-approximate for a sufficiently small constant δ>0\delta>0. We also refute the existence of deterministic o(n)o(n)-query O(logn)O(\log n)-approximation algorithms.

Keywords

Cite

@article{arxiv.2201.01436,
  title  = {Deterministic metric $1$-median selection with very few queries},
  author = {Ching-Lueh Chang},
  journal= {arXiv preprint arXiv:2201.01436},
  year   = {2022}
}

Comments

Part of this paper appears in {\em Proceedings of the 27th International Computing and Combinatorics Conference} (COCOON 2021)

R2 v1 2026-06-24T08:40:29.693Z