Inferring Structure and Parameters of Dynamic Systems using Latin Hypercube Sampling Multi Dimensional Uniformity-Particle Swarm Optimization

Muhammad Usman, Abubakr Awad, Wei Pang, George M. Coghill

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Inferring models of dynamic systems from their time series data is a challenging task for optimization algorithms due to its potentially expensive computational cost and underlying large search space. In this study, we aim to infer both the structure and parameters of a dynamic system model simultaneously by Particle Swarm Optimization (PSO), enhanced by effective stratified sampling strategies.
More specifically, we apply Latin Hyper Cube Sampling (LHS) with PSO. This leads to two novel swarm-inspired algorithms, LHS-PSO which can be used efficiently to learn the structure and parameters of simple and complex dynamic system models. We used a complex biological cancer model called Kinetochores, for assessing the performance of PSO and LHS-PSO. The experimental results
demonstrate that LHS-PSO can find promising solutions with corresponding structure and parameters, and it outperforms PSO during our experiments.
Original languageEnglish
Title of host publicationGECCO 2019 Companion - Proceedings of the 2019 Genetic and Evolutionary Computation Conference Companion
EditorsManuel López-Ibáñez, Anne Auger, Thomas Stützle
Place of PublicationNew York, USA
PublisherACM
Pages101-102
Number of pages2
ISBN (Electronic)9781450367486
ISBN (Print)9781450367486
DOIs
Publication statusPublished - 13 Jul 2019
EventThe Genetic and Evolutionary Computation Conference GECCO 2019 - Prague, Czech Republic
Duration: 13 Jul 201917 Jul 2019

Conference

ConferenceThe Genetic and Evolutionary Computation Conference GECCO 2019
CountryCzech Republic
CityPrague
Period13/07/1917/07/19

Fingerprint

Particle swarm optimization (PSO)
Dynamical systems
Sampling
Time series
Costs
Experiments

Keywords

  • Dynamic Systems
  • Particle Swarm Optimization
  • Genetic Algorithm
  • Latin Hypercube Sampling
  • Learning Structure and Parameter
  • Parameter
  • Learning Structure

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence
  • Theoretical Computer Science

Cite this

Usman, M., Awad, A., Pang, W., & Coghill, G. M. (2019). Inferring Structure and Parameters of Dynamic Systems using Latin Hypercube Sampling Multi Dimensional Uniformity-Particle Swarm Optimization. In M. López-Ibáñez, A. Auger, & T. Stützle (Eds.), GECCO 2019 Companion - Proceedings of the 2019 Genetic and Evolutionary Computation Conference Companion (pp. 101-102). New York, USA: ACM. https://doi.org/10.1145/3319619.3322010

Inferring Structure and Parameters of Dynamic Systems using Latin Hypercube Sampling Multi Dimensional Uniformity-Particle Swarm Optimization. / Usman, Muhammad; Awad, Abubakr; Pang, Wei; Coghill, George M.

GECCO 2019 Companion - Proceedings of the 2019 Genetic and Evolutionary Computation Conference Companion. ed. / Manuel López-Ibáñez; Anne Auger; Thomas Stützle. New York, USA : ACM, 2019. p. 101-102.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Usman, M, Awad, A, Pang, W & Coghill, GM 2019, Inferring Structure and Parameters of Dynamic Systems using Latin Hypercube Sampling Multi Dimensional Uniformity-Particle Swarm Optimization. in M López-Ibáñez, A Auger & T Stützle (eds), GECCO 2019 Companion - Proceedings of the 2019 Genetic and Evolutionary Computation Conference Companion. ACM, New York, USA, pp. 101-102, The Genetic and Evolutionary Computation Conference GECCO 2019, Prague, Czech Republic, 13/07/19. https://doi.org/10.1145/3319619.3322010
Usman M, Awad A, Pang W, Coghill GM. Inferring Structure and Parameters of Dynamic Systems using Latin Hypercube Sampling Multi Dimensional Uniformity-Particle Swarm Optimization. In López-Ibáñez M, Auger A, Stützle T, editors, GECCO 2019 Companion - Proceedings of the 2019 Genetic and Evolutionary Computation Conference Companion. New York, USA: ACM. 2019. p. 101-102 https://doi.org/10.1145/3319619.3322010
Usman, Muhammad ; Awad, Abubakr ; Pang, Wei ; Coghill, George M. / Inferring Structure and Parameters of Dynamic Systems using Latin Hypercube Sampling Multi Dimensional Uniformity-Particle Swarm Optimization. GECCO 2019 Companion - Proceedings of the 2019 Genetic and Evolutionary Computation Conference Companion. editor / Manuel López-Ibáñez ; Anne Auger ; Thomas Stützle. New York, USA : ACM, 2019. pp. 101-102
@inproceedings{bebdf2abf51f4e028cd3a9ded8fd817c,
title = "Inferring Structure and Parameters of Dynamic Systems using Latin Hypercube Sampling Multi Dimensional Uniformity-Particle Swarm Optimization",
abstract = "Inferring models of dynamic systems from their time series data is a challenging task for optimization algorithms due to its potentially expensive computational cost and underlying large search space. In this study, we aim to infer both the structure and parameters of a dynamic system model simultaneously by Particle Swarm Optimization (PSO), enhanced by effective stratified sampling strategies.More specifically, we apply Latin Hyper Cube Sampling (LHS) with PSO. This leads to two novel swarm-inspired algorithms, LHS-PSO which can be used efficiently to learn the structure and parameters of simple and complex dynamic system models. We used a complex biological cancer model called Kinetochores, for assessing the performance of PSO and LHS-PSO. The experimental resultsdemonstrate that LHS-PSO can find promising solutions with corresponding structure and parameters, and it outperforms PSO during our experiments.",
keywords = "Dynamic Systems, Particle Swarm Optimization, Genetic Algorithm, Latin Hypercube Sampling, Learning Structure and Parameter, Parameter, Learning Structure",
author = "Muhammad Usman and Abubakr Awad and Wei Pang and Coghill, {George M.}",
note = "Muhammad Usman and Abubakr Awad are supported by Elphinstone PhD Scholarship (University of Aberdeen).",
year = "2019",
month = "7",
day = "13",
doi = "10.1145/3319619.3322010",
language = "English",
isbn = "9781450367486",
pages = "101--102",
editor = "Manuel L{\'o}pez-Ib{\'a}{\~n}ez and Anne Auger and Thomas St{\"u}tzle",
booktitle = "GECCO 2019 Companion - Proceedings of the 2019 Genetic and Evolutionary Computation Conference Companion",
publisher = "ACM",

}

TY - GEN

T1 - Inferring Structure and Parameters of Dynamic Systems using Latin Hypercube Sampling Multi Dimensional Uniformity-Particle Swarm Optimization

AU - Usman, Muhammad

AU - Awad, Abubakr

AU - Pang, Wei

AU - Coghill, George M.

N1 - Muhammad Usman and Abubakr Awad are supported by Elphinstone PhD Scholarship (University of Aberdeen).

PY - 2019/7/13

Y1 - 2019/7/13

N2 - Inferring models of dynamic systems from their time series data is a challenging task for optimization algorithms due to its potentially expensive computational cost and underlying large search space. In this study, we aim to infer both the structure and parameters of a dynamic system model simultaneously by Particle Swarm Optimization (PSO), enhanced by effective stratified sampling strategies.More specifically, we apply Latin Hyper Cube Sampling (LHS) with PSO. This leads to two novel swarm-inspired algorithms, LHS-PSO which can be used efficiently to learn the structure and parameters of simple and complex dynamic system models. We used a complex biological cancer model called Kinetochores, for assessing the performance of PSO and LHS-PSO. The experimental resultsdemonstrate that LHS-PSO can find promising solutions with corresponding structure and parameters, and it outperforms PSO during our experiments.

AB - Inferring models of dynamic systems from their time series data is a challenging task for optimization algorithms due to its potentially expensive computational cost and underlying large search space. In this study, we aim to infer both the structure and parameters of a dynamic system model simultaneously by Particle Swarm Optimization (PSO), enhanced by effective stratified sampling strategies.More specifically, we apply Latin Hyper Cube Sampling (LHS) with PSO. This leads to two novel swarm-inspired algorithms, LHS-PSO which can be used efficiently to learn the structure and parameters of simple and complex dynamic system models. We used a complex biological cancer model called Kinetochores, for assessing the performance of PSO and LHS-PSO. The experimental resultsdemonstrate that LHS-PSO can find promising solutions with corresponding structure and parameters, and it outperforms PSO during our experiments.

KW - Dynamic Systems

KW - Particle Swarm Optimization

KW - Genetic Algorithm

KW - Latin Hypercube Sampling

KW - Learning Structure and Parameter

KW - Parameter

KW - Learning Structure

UR - http://www.mendeley.com/research/inferring-structure-parameters-dynamic-systems-using-particle-swarm-optimization

UR - http://www.scopus.com/inward/record.url?scp=85070601203&partnerID=8YFLogxK

UR - https://abdn.pure.elsevier.com/en/en/researchoutput/inferring-structure-and-parameters-of-dynamic-systems-using-latin-hypercube-sampling-multi-dimensional-uniformityparticle-swarm-optimization(bebdf2ab-f51f-4e02-8cd3-a9ded8fd817c).html

UR - http://pure-ots.abdn.ac.uk:8080/portal/en/researchoutput/inferring-structure-and-parameters-of-dynamic-systems-using-latin-hypercube-sampling-multi-dimensional-uniformityparticle-swarm-optimization(bebdf2ab-f51f-4e02-8cd3-a9ded8fd817c).html

U2 - 10.1145/3319619.3322010

DO - 10.1145/3319619.3322010

M3 - Conference contribution

SN - 9781450367486

SP - 101

EP - 102

BT - GECCO 2019 Companion - Proceedings of the 2019 Genetic and Evolutionary Computation Conference Companion

A2 - López-Ibáñez, Manuel

A2 - Auger, Anne

A2 - Stützle, Thomas

PB - ACM

CY - New York, USA

ER -