Surfing on Fitness Landscapes: A Boost on Optimization by Fourier Surrogate Modeling
Surfing in rough waters is not always as fun as wave riding the “big one”. Similarly, in optimization problems, fitness landscapes with a huge number of local optima make the search for the global optimum a hard and generally annoying game. Computational Intelligence optimization...
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2020-02-01
|
Series: | Entropy |
Subjects: | |
Online Access: | https://www.mdpi.com/1099-4300/22/3/285 |
_version_ | 1811307854072643584 |
---|---|
author | Luca Manzoni Daniele M. Papetti Paolo Cazzaniga Simone Spolaor Giancarlo Mauri Daniela Besozzi Marco S. Nobile |
author_facet | Luca Manzoni Daniele M. Papetti Paolo Cazzaniga Simone Spolaor Giancarlo Mauri Daniela Besozzi Marco S. Nobile |
author_sort | Luca Manzoni |
collection | DOAJ |
description | Surfing in rough waters is not always as fun as wave riding the “big one”. Similarly, in optimization problems, fitness landscapes with a huge number of local optima make the search for the global optimum a hard and generally annoying game. Computational Intelligence optimization metaheuristics use a set of individuals that “surf” across the fitness landscape, sharing and exploiting pieces of information about local fitness values in a joint effort to find out the global optimum. In this context, we designed surF, a novel surrogate modeling technique that leverages the discrete Fourier transform to generate a smoother, and possibly easier to explore, fitness landscape. The rationale behind this idea is that filtering out the high frequencies of the fitness function and keeping only its partial information (i.e., the low frequencies) can actually be beneficial in the optimization process. We prove our theory by combining surF with a settings free variant of Particle Swarm Optimization (PSO) based on Fuzzy Logic, called Fuzzy Self-Tuning PSO. Specifically, we introduce a new algorithm, named F3ST-PSO, which performs a preliminary exploration on the surrogate model followed by a second optimization using the actual fitness function. We show that F3ST-PSO can lead to improved performances, notably using the same budget of fitness evaluations. |
first_indexed | 2024-04-13T09:11:44Z |
format | Article |
id | doaj.art-eec0c10091d44e2b8a149676fded7639 |
institution | Directory Open Access Journal |
issn | 1099-4300 |
language | English |
last_indexed | 2024-04-13T09:11:44Z |
publishDate | 2020-02-01 |
publisher | MDPI AG |
record_format | Article |
series | Entropy |
spelling | doaj.art-eec0c10091d44e2b8a149676fded76392022-12-22T02:52:52ZengMDPI AGEntropy1099-43002020-02-0122328510.3390/e22030285e22030285Surfing on Fitness Landscapes: A Boost on Optimization by Fourier Surrogate ModelingLuca Manzoni0Daniele M. Papetti1Paolo Cazzaniga2Simone Spolaor3Giancarlo Mauri4Daniela Besozzi5Marco S. Nobile6Department of Mathematics and Geosciences, University of Trieste, 34127 Trieste, ItalyDepartment of Informatics, Systems and Communication, University of Milano-Bicocca, 20126 Milano, ItalyDepartment of Human and Social Sciences, University of Bergamo, 24129 Bergamo, ItalyDepartment of Informatics, Systems and Communication, University of Milano-Bicocca, 20126 Milano, ItalyDepartment of Informatics, Systems and Communication, University of Milano-Bicocca, 20126 Milano, ItalyDepartment of Informatics, Systems and Communication, University of Milano-Bicocca, 20126 Milano, ItalyDepartment of Industrial Engineering & Innovation Sciences, Eindhoven University of Technology, 5612 AZ Eindhoven, The NetherlandsSurfing in rough waters is not always as fun as wave riding the “big one”. Similarly, in optimization problems, fitness landscapes with a huge number of local optima make the search for the global optimum a hard and generally annoying game. Computational Intelligence optimization metaheuristics use a set of individuals that “surf” across the fitness landscape, sharing and exploiting pieces of information about local fitness values in a joint effort to find out the global optimum. In this context, we designed surF, a novel surrogate modeling technique that leverages the discrete Fourier transform to generate a smoother, and possibly easier to explore, fitness landscape. The rationale behind this idea is that filtering out the high frequencies of the fitness function and keeping only its partial information (i.e., the low frequencies) can actually be beneficial in the optimization process. We prove our theory by combining surF with a settings free variant of Particle Swarm Optimization (PSO) based on Fuzzy Logic, called Fuzzy Self-Tuning PSO. Specifically, we introduce a new algorithm, named F3ST-PSO, which performs a preliminary exploration on the surrogate model followed by a second optimization using the actual fitness function. We show that F3ST-PSO can lead to improved performances, notably using the same budget of fitness evaluations.https://www.mdpi.com/1099-4300/22/3/285global optimizationparticle swarm optimizationfuzzy self-tuning psofourier transformsurrogate modeling |
spellingShingle | Luca Manzoni Daniele M. Papetti Paolo Cazzaniga Simone Spolaor Giancarlo Mauri Daniela Besozzi Marco S. Nobile Surfing on Fitness Landscapes: A Boost on Optimization by Fourier Surrogate Modeling Entropy global optimization particle swarm optimization fuzzy self-tuning pso fourier transform surrogate modeling |
title | Surfing on Fitness Landscapes: A Boost on Optimization by Fourier Surrogate Modeling |
title_full | Surfing on Fitness Landscapes: A Boost on Optimization by Fourier Surrogate Modeling |
title_fullStr | Surfing on Fitness Landscapes: A Boost on Optimization by Fourier Surrogate Modeling |
title_full_unstemmed | Surfing on Fitness Landscapes: A Boost on Optimization by Fourier Surrogate Modeling |
title_short | Surfing on Fitness Landscapes: A Boost on Optimization by Fourier Surrogate Modeling |
title_sort | surfing on fitness landscapes a boost on optimization by fourier surrogate modeling |
topic | global optimization particle swarm optimization fuzzy self-tuning pso fourier transform surrogate modeling |
url | https://www.mdpi.com/1099-4300/22/3/285 |
work_keys_str_mv | AT lucamanzoni surfingonfitnesslandscapesaboostonoptimizationbyfouriersurrogatemodeling AT danielempapetti surfingonfitnesslandscapesaboostonoptimizationbyfouriersurrogatemodeling AT paolocazzaniga surfingonfitnesslandscapesaboostonoptimizationbyfouriersurrogatemodeling AT simonespolaor surfingonfitnesslandscapesaboostonoptimizationbyfouriersurrogatemodeling AT giancarlomauri surfingonfitnesslandscapesaboostonoptimizationbyfouriersurrogatemodeling AT danielabesozzi surfingonfitnesslandscapesaboostonoptimizationbyfouriersurrogatemodeling AT marcosnobile surfingonfitnesslandscapesaboostonoptimizationbyfouriersurrogatemodeling |