104
7

Local approximation algorithms for a class of 0/1 max-min linear programs

Abstract

We study the applicability of distributed, local algorithms to 0/1 max-min LPs where the objective is to maximise minkvckvxv{\min_k \sum_v c_{kv} x_v} subject to vaivxv1{\sum_v a_{iv} x_v \le 1} for each ii and xv0{x_v \ge 0} for each vv. Here ckv{0,1}c_{kv} \in \{0,1\}, aiv{0,1}a_{iv} \in \{0,1\}, and the support sets Vi={v:aiv>0}{V_i = \{v : a_{iv} > 0 \}} and Vk={v:ckv>0}{V_k = \{v : c_{kv}>0 \}} have bounded size; in particular, we study the case Vk2|V_k| \le 2. Each agent vv is responsible for choosing the value of xvx_v based on information within its constant-size neighbourhood; the communication network is the hypergraph where the sets VkV_k and ViV_i constitute the hyperedges. We present a local approximation algorithm which achieves an approximation ratio arbitrarily close to the theoretical lower bound presented in prior work.

View on arXiv
Comments on this paper