Abstract
In this paper the influence of structural information obtained from a problem relaxation on the performance of an ACO algorithm for the symmetric TSP is studied. More precisely, a very simple ACO algorithm is guided by including Minimal Spanning Tree information into the visibility. Empirical results on a large number of benchmark instances from TSPLIB are presented. The paper concludes with remarks on some more elaborate ideas for using problem relaxation within ACO. © Springer-Verlag Berlin Heidelberg 2007.
Cite
CITATION STYLE
Reimann, M. (2007). Guiding ACO by problem relaxation: A case study on the symmetric TSP. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4771 LNCS, pp. 45–55). Springer Verlag. https://doi.org/10.1007/978-3-540-75514-2_4
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
