The impact of particles initialization in PSO: parameter estimation as a case in point

Paolo Cazzaniga; Marco S. Nobile; Daniela Besozzi

doi:10.1109/CIBCB.2015.7300288

The impact of particles initialization in PSO: parameter estimation as a case in point

Paolo Cazzaniga, Marco S. Nobile, Daniela Besozzi

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

43 Citations (Scopus)

Abstract

Despite the intense research focused on the investigation of the functioning settings of Particle Swarm Optimization, the particles initialization functions - determining the initial positions in the search space - are generally ignored, especially in the case of real-world applications. As a matter of fact, almost all works exploit uniform distributions to randomly generate the particles coordinates. In this article, we analyze the impact on the optimization performances of alternative initialization functions based on logarithmic, normal, and lognormal distributions. Our results show how different initialization strategies can affect - and in some cases largely improve - the convergence speed, both in the case of benchmark functions and in the optimization of the kinetic constants of biochemical systems.

Original language	English
Title of host publication	2015 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology, CIBCB 2015
Place of Publication	Piscataway
Publisher	Institute of Electrical and Electronics Engineers
Number of pages	8
ISBN (Electronic)	978-1-4799-6926-5
DOIs	https://doi.org/10.1109/CIBCB.2015.7300288
Publication status	Published - 16 Oct 2015
Externally published	Yes
Event	2015 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology, CIBCB 2015 - Niagara Falls, Canada Duration: 12 Aug 2015 → 15 Aug 2015

Conference

Conference	2015 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology, CIBCB 2015
Country/Territory	Canada
City	Niagara Falls
Period	12/08/15 → 15/08/15

Access to Document

10.1109/CIBCB.2015.7300288

Cite this

@inproceedings{10a2328976a3451080dac5c2e9f5b49e,

title = "The impact of particles initialization in PSO: parameter estimation as a case in point",

abstract = "Despite the intense research focused on the investigation of the functioning settings of Particle Swarm Optimization, the particles initialization functions - determining the initial positions in the search space - are generally ignored, especially in the case of real-world applications. As a matter of fact, almost all works exploit uniform distributions to randomly generate the particles coordinates. In this article, we analyze the impact on the optimization performances of alternative initialization functions based on logarithmic, normal, and lognormal distributions. Our results show how different initialization strategies can affect - and in some cases largely improve - the convergence speed, both in the case of benchmark functions and in the optimization of the kinetic constants of biochemical systems.",

author = "Paolo Cazzaniga and Nobile, {Marco S.} and Daniela Besozzi",

year = "2015",

month = oct,

day = "16",

doi = "10.1109/CIBCB.2015.7300288",

language = "English",

booktitle = "2015 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology, CIBCB 2015",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

note = "2015 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology, CIBCB 2015 ; Conference date: 12-08-2015 Through 15-08-2015",

}

Cazzaniga, P, Nobile, MS & Besozzi, D 2015, The impact of particles initialization in PSO: parameter estimation as a case in point. in 2015 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology, CIBCB 2015., 7300288, Institute of Electrical and Electronics Engineers, Piscataway, 2015 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology, CIBCB 2015, Niagara Falls, Canada, 12/08/15. https://doi.org/10.1109/CIBCB.2015.7300288

The impact of particles initialization in PSO: parameter estimation as a case in point. / Cazzaniga, Paolo; Nobile, Marco S.; Besozzi, Daniela.
2015 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology, CIBCB 2015. Piscataway: Institute of Electrical and Electronics Engineers, 2015. 7300288.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - The impact of particles initialization in PSO

T2 - 2015 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology, CIBCB 2015

AU - Cazzaniga, Paolo

AU - Nobile, Marco S.

AU - Besozzi, Daniela

PY - 2015/10/16

Y1 - 2015/10/16

N2 - Despite the intense research focused on the investigation of the functioning settings of Particle Swarm Optimization, the particles initialization functions - determining the initial positions in the search space - are generally ignored, especially in the case of real-world applications. As a matter of fact, almost all works exploit uniform distributions to randomly generate the particles coordinates. In this article, we analyze the impact on the optimization performances of alternative initialization functions based on logarithmic, normal, and lognormal distributions. Our results show how different initialization strategies can affect - and in some cases largely improve - the convergence speed, both in the case of benchmark functions and in the optimization of the kinetic constants of biochemical systems.

AB - Despite the intense research focused on the investigation of the functioning settings of Particle Swarm Optimization, the particles initialization functions - determining the initial positions in the search space - are generally ignored, especially in the case of real-world applications. As a matter of fact, almost all works exploit uniform distributions to randomly generate the particles coordinates. In this article, we analyze the impact on the optimization performances of alternative initialization functions based on logarithmic, normal, and lognormal distributions. Our results show how different initialization strategies can affect - and in some cases largely improve - the convergence speed, both in the case of benchmark functions and in the optimization of the kinetic constants of biochemical systems.

UR - http://www.scopus.com/inward/record.url?scp=84953432811&partnerID=8YFLogxK

U2 - 10.1109/CIBCB.2015.7300288

DO - 10.1109/CIBCB.2015.7300288

M3 - Conference contribution

AN - SCOPUS:84953432811

BT - 2015 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology, CIBCB 2015

PB - Institute of Electrical and Electronics Engineers

CY - Piscataway

Y2 - 12 August 2015 through 15 August 2015

ER -

The impact of particles initialization in PSO: parameter estimation as a case in point

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this