Year
2000
Authors
ALFANDARI Laurent, PASCHOS V.
Abstract
Many minimization problems consist in covering vertices or edges of a graph by subgraphs verifying a given property. If the number of candidate subgraphs is a polynomial function depending on the size of the graph, then the classical greedy heuristic for Set Covering, analysed by Chvatal, guarantees a polynomial approximation within a logarithmic ratio. We extend this result to a class of problems with a possible exponential number of subgraphs, by revisiting a technique called “master-slave”. Applications are studied for telecommunication, optimal location and VLSI-design.
ALFANDARI, L. et PASCHOS, V. (2000). Master-slave Strategy and Polynomial Approximation. Computational Optimization and Applications, pp. 231-245.