Data From "Clarifying the Difference in Local Optima Network Sampling Algorithms"

Abstract

The dataset comprises fitness landscape data for "Clarifying the Differences in Local Optima Network Sampling Algorithms", S. L. Thomson, G. Ochoa and S. Verel, The 19th European Conference on Evolutionary Computation in Combinatorial Optimisation (EvoCOP, part of EvoStar 2019), 24-26 April 2019, Leipzig, Germany. The dataset is extracted local optima networks for the 30 Quadratic Assignment Problem instances that are sampled in the paper according to two different methodologies: using a snowball sampling approach, and using an iterated local search based algorithm. Also included is optimisation data for the instances, obtained by running iterated local search and robust tabu search on them multiple times.