CUSTOMHyS: Customising Optimisation Metaheuristics via Hyper-heuristic Search
Jorge M. Cruz-Duarte,
Ivan Amaya,
José C. Ortiz-Bayliss,
Hugo Terashima-Marín,
Yong Shi
Affiliations
Jorge M. Cruz-Duarte
School of Engineering and Sciences, Tecnologico de Monterrey, Av. Eugenio Garza Sada 2501 Sur, Monterrey, NL 64849, Mexico
Ivan Amaya
School of Engineering and Sciences, Tecnologico de Monterrey, Av. Eugenio Garza Sada 2501 Sur, Monterrey, NL 64849, Mexico; Corresponding author.
José C. Ortiz-Bayliss
School of Engineering and Sciences, Tecnologico de Monterrey, Av. Eugenio Garza Sada 2501 Sur, Monterrey, NL 64849, Mexico
Hugo Terashima-Marín
School of Engineering and Sciences, Tecnologico de Monterrey, Av. Eugenio Garza Sada 2501 Sur, Monterrey, NL 64849, Mexico
Yong Shi
Research Center on Fictitious Economy and Data Science, Chinese Academy of Sciences, Zhongguancun East Road 80, Haidian District, Beijing 100190, China
There is a colourful palette of metaheuristics for solving continuous optimisation problems in the literature. Unfortunately, it is not easy to pick a suitable one for a specific practical scenario. Moreover, oftentimes the selected metaheuristic must be tuned until finding adequate parameter settings. Therefore, this work presents a framework based on a hyper-heuristic powered by Simulated Annealing for tailoring population-based metaheuristics. To do so, we recognise search operators from well-known techniques as building blocks for new ones. The presented framework comprises six main modules coded in Python, which can be used independently, and which help explore new metaheuristics.
