How crossover speeds up evolutionary algorithms for the multi-criteria all-pairs-shortest-path problem

27Citations
Citations of this article
12Readers
Mendeley users who have this article in their library.
Get full text

Abstract

Understanding the impact of crossover in evolutionary algorithms is one of the major challenges in the theoretical analysis of these stochastic search algorithms. Recently, it has been shown that crossover provably helps to speed up evolutionary algorithms for the classical all-pairs-shortest path (APSP) problem. In this paper, we extend this approach to the NP-hard multi-criteria APSP problem. Based on rigorous runtime analyses, we point out that crossover leads to better worst case bounds than previous known results. This is the first time that rigorous runtime analyses have shown the usefulness of crossover for an NP-hard multi-criteria optimization problem. © 2010 Springer-Verlag.

Cite

CITATION STYLE

APA

Neumann, F., & Theile, M. (2010). How crossover speeds up evolutionary algorithms for the multi-criteria all-pairs-shortest-path problem. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6238 LNCS, pp. 667–676). https://doi.org/10.1007/978-3-642-15844-5_67

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free