Landscape is an important notion in studying the difficulty of a combinatorial problem and the behavior of heuristics. In this pa- per, two new measures for analyzing landscapes are introduced, each of them based on the Hamming distance of iso-cost levels. Sampling techniques based on neighborhood search are defined in order to effect an approximation of these measures. These measures and techniques are used to analyze and characterize the properties of Maximal Constraint Satisfaction Problem random landscapes.
CITATION STYLE
Belaidouni, M., & Hao, J. K. (2000). Landscapes and the maximal constraint satisfaction problem. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1829, pp. 242–253). Springer Verlag. https://doi.org/10.1007/10721187_18
Mendeley helps you to discover research relevant for your work.