Raw data & extended results for "HyperSPAM: A study on hyper-heuristic Coordination Strategies in the Continuous Domain"

Published: 9 June 2019| Version 1 | DOI: 10.17632/6st7grtxfr.1
Contributors:
Fabio Caraffini, Ferrante Neri, Michael G. Epitropakis

Description

This dataset contains a pdf file with an extended list of tables and images that could not be included in the main paper. For details on the experimental setup, parameters tuning and performed statistical analysis please refer to the published paper. Moreover, numerical data produced to carry out this investigation are made available for reproducibility and dissemination reasons. The "ComparisonAgainstStateOfTheArt" folder contains: - a zip archive with data of the four presented hyperSPAM variants as well as those for three state-of-the-art algorithms over popular benchmark problems at multiple dimensionality values; - a zip archive with data of the best hyperSPAM variant as well as those for three state-of-the-art algorithms over three real-world applications from CEC2011. Similarly, the "hyperSPAM-Subcomponents" folder contains: - a zip archive file with data of three "degenerative" hyperSPAM algorithms obtained by deactivating memes, as described in the published paper, ran over the same benchmark problems; - a zip archive file with data obtained with such sub-variants over the same real-world applications from CEC 2011.

Files

Institutions

  • De Montfort University

Categories

Heuristics, Continuous Optimization, Memetic Algorithm, Automated Computational Design, Computer Aspects of Numerical Algorithms

Licence