TY - JOUR
T1 - Simplifying Fitness Landscapes Using Dilation Functions Evolved With Genetic Programming
AU - Papetti, Daniele M.
AU - Tangherloni, Andrea
AU - Farinati, Davide
AU - Cazzaniga, Paolo
AU - Vanneschi, Leonardo
N1 - info:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UIDB%2F04152%2F2020/PT#
Papetti, D. M., Tangherloni, A., Farinati, D., Cazzaniga, P., & Vanneschi, L. (2023). Simplifying Fitness Landscapes Using Dilation Functions Evolved With Genetic Programming. IEEE Computational Intelligence Magazine, 18(1), 22-31. https://doi.org/10.1109/MCI.2022.3222096. --- Funding Information: This work was supported by the National Funds through FCT (Fundão para a Ciencia e a Tecnologia, Portugal), under Project - UIDB/04152/2020 - Centro de Investigão em Gestão de Informão (MagIC)/NOVA IMS. This article has supplementary downloadable material available at https://doi.org/10.1109/MCI.2022.3222096, provided by the authors.
PY - 2023/2/1
Y1 - 2023/2/1
N2 - Several optimization problems have features that hinder the capabilities of searching heuristics. To cope with this issue, different methods have been proposed to manipulate search spaces and improve the optimization process. This paper focuses on Dilation Functions (DFs), which are one of the most promising techniques to manipulate the fitness landscape, by “expanding” or “compressing” specific regions. The definition of appropriate DFs is problem dependent and requires a-priori knowledge of the optimization problem. Therefore, it is essential to introduce an automatic and efficient strategy to identify optimal DFs. With this aim, we propose a novel method based on Genetic Programming, named GP4DFs, which is capable of evolving effective DFs. GP4DFs identifies optimal dilations, where a specific DF is applied to each dimension of the search space. Moreover, thanks to a knowledge-driven initialization strategy, GP4DFs converges to better solutions with a reduced number of fitness evaluations, compared to the state-of-the-art approaches. The performance of GP4DFs is assessed on a set of 43 benchmark functions mimicking several features of real-world optimization problems. The obtained results indicate the suitability of the generated DFs.
AB - Several optimization problems have features that hinder the capabilities of searching heuristics. To cope with this issue, different methods have been proposed to manipulate search spaces and improve the optimization process. This paper focuses on Dilation Functions (DFs), which are one of the most promising techniques to manipulate the fitness landscape, by “expanding” or “compressing” specific regions. The definition of appropriate DFs is problem dependent and requires a-priori knowledge of the optimization problem. Therefore, it is essential to introduce an automatic and efficient strategy to identify optimal DFs. With this aim, we propose a novel method based on Genetic Programming, named GP4DFs, which is capable of evolving effective DFs. GP4DFs identifies optimal dilations, where a specific DF is applied to each dimension of the search space. Moreover, thanks to a knowledge-driven initialization strategy, GP4DFs converges to better solutions with a reduced number of fitness evaluations, compared to the state-of-the-art approaches. The performance of GP4DFs is assessed on a set of 43 benchmark functions mimicking several features of real-world optimization problems. The obtained results indicate the suitability of the generated DFs.
KW - Evolution (biology)
KW - Sociology
KW - Genetic programming
KW - Transforms
KW - Benchmark testing
KW - Search problems
KW - Statistics
UR - https://doi.org/10.1109/MCI.2022.3222096/mm1
UR - http://www.scopus.com/inward/record.url?scp=85147731150&partnerID=8YFLogxK
UR - https://www.webofscience.com/wos/woscc/full-record/WOS:000922860500007
U2 - 10.1109/MCI.2022.3222096
DO - 10.1109/MCI.2022.3222096
M3 - Article
VL - 18
SP - 22
EP - 31
JO - IEEE Computational Intelligence Magazine
JF - IEEE Computational Intelligence Magazine
SN - 1556-603X
IS - 1
ER -