Approximation Algorithms for Maximum Coverage and Max Cut with Given Sizes of Parts

Alexander A. Ageev, Maxim Sviridenko

Abstract: In this paper we demonstrate a general method of designing constant-factor approximation algorithms for some discrete optimization problems with cardinality constraints. The core of the method is a simple deterministic (“pipage”) procedure of rounding of linear relaxations. By using the method we design a (1 − (1 − 1/k)k)-approximation algorithm for the maximum coverage problem where k is the maximum size of the subsets that are covered, and a 1/2-approximation algorithm for the maximum cut problem with given sizes of parts in the vertex set bipartition. The performance guarantee of the former improves on that of the well-known (1 − e −1)-greedy algorithm due to Cornuejols, Fisher and Nemhauser in each case of bounded k. The latter is, to the best of our knowledge, the first constant-factor algorithm for that version of the maximum cut problem.