Application of on-line machine learning in optimization algorithms: A case study for local search

Cong Hao, Takeshi Yoshimura

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

Abstract

The study on machine learning has been flourishing for several years, and machine learning algorithms are being applied to various fields with great achievements. In this paper, combining the on-line machine learning method into optimization algorithms is to be studied. In many heuristic optimization algorithms, one common way to reduce execution time and improve solution optimality is, first estimating the quality of a set of candidate solutions, and solving only promising candidates in detail. Currently most estimations are performed by empirical equations, whose accuracy greatly relies on the how well the equation is designed. In this paper, we propose an on-line learning based estimator to perform the solution estimation in heuristic algorithms to improve estimation accuracy. Then a simple case study is discussed, where a local search based heuristic with random start is used, and an on-line estimator considering the properties of local search is proposed. The experiments show that the accuracy of on-line estimator is much higher than the static estimator, and is also higher than a general off-line pre-Trained learner. Even though the on-line estimator introduced special time for its training, the heuristic algorithm still speeds up by 3.7X without optimality sacrifice.

Original languageEnglish
Title of host publication2017 9th Computer Science and Electronic Engineering Conference, CEEC 2017 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages19-24
Number of pages6
ISBN (Electronic)9781538630075
DOIs
Publication statusPublished - 2017 Nov 8
Event9th Computer Science and Electronic Engineering Conference, CEEC 2017 - Colchester, United Kingdom
Duration: 2017 Sep 272017 Sep 29

Other

Other9th Computer Science and Electronic Engineering Conference, CEEC 2017
CountryUnited Kingdom
CityColchester
Period17/9/2717/9/29

Fingerprint

Learning systems
Heuristic algorithms
Learning algorithms
Local search (optimization)
Experiments

ASJC Scopus subject areas

  • Computer Science (miscellaneous)
  • Computer Networks and Communications
  • Computer Science Applications
  • Electrical and Electronic Engineering

Cite this

Hao, C., & Yoshimura, T. (2017). Application of on-line machine learning in optimization algorithms: A case study for local search. In 2017 9th Computer Science and Electronic Engineering Conference, CEEC 2017 - Proceedings (pp. 19-24). [8101593] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CEEC.2017.8101593

Application of on-line machine learning in optimization algorithms : A case study for local search. / Hao, Cong; Yoshimura, Takeshi.

2017 9th Computer Science and Electronic Engineering Conference, CEEC 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. p. 19-24 8101593.

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

Hao, C & Yoshimura, T 2017, Application of on-line machine learning in optimization algorithms: A case study for local search. in 2017 9th Computer Science and Electronic Engineering Conference, CEEC 2017 - Proceedings., 8101593, Institute of Electrical and Electronics Engineers Inc., pp. 19-24, 9th Computer Science and Electronic Engineering Conference, CEEC 2017, Colchester, United Kingdom, 17/9/27. https://doi.org/10.1109/CEEC.2017.8101593
Hao C, Yoshimura T. Application of on-line machine learning in optimization algorithms: A case study for local search. In 2017 9th Computer Science and Electronic Engineering Conference, CEEC 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2017. p. 19-24. 8101593 https://doi.org/10.1109/CEEC.2017.8101593
Hao, Cong ; Yoshimura, Takeshi. / Application of on-line machine learning in optimization algorithms : A case study for local search. 2017 9th Computer Science and Electronic Engineering Conference, CEEC 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 19-24
@inproceedings{b7aed06e5a604b639040c5a3ee375b34,
title = "Application of on-line machine learning in optimization algorithms: A case study for local search",
abstract = "The study on machine learning has been flourishing for several years, and machine learning algorithms are being applied to various fields with great achievements. In this paper, combining the on-line machine learning method into optimization algorithms is to be studied. In many heuristic optimization algorithms, one common way to reduce execution time and improve solution optimality is, first estimating the quality of a set of candidate solutions, and solving only promising candidates in detail. Currently most estimations are performed by empirical equations, whose accuracy greatly relies on the how well the equation is designed. In this paper, we propose an on-line learning based estimator to perform the solution estimation in heuristic algorithms to improve estimation accuracy. Then a simple case study is discussed, where a local search based heuristic with random start is used, and an on-line estimator considering the properties of local search is proposed. The experiments show that the accuracy of on-line estimator is much higher than the static estimator, and is also higher than a general off-line pre-Trained learner. Even though the on-line estimator introduced special time for its training, the heuristic algorithm still speeds up by 3.7X without optimality sacrifice.",
author = "Cong Hao and Takeshi Yoshimura",
year = "2017",
month = "11",
day = "8",
doi = "10.1109/CEEC.2017.8101593",
language = "English",
pages = "19--24",
booktitle = "2017 9th Computer Science and Electronic Engineering Conference, CEEC 2017 - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Application of on-line machine learning in optimization algorithms

T2 - A case study for local search

AU - Hao, Cong

AU - Yoshimura, Takeshi

PY - 2017/11/8

Y1 - 2017/11/8

N2 - The study on machine learning has been flourishing for several years, and machine learning algorithms are being applied to various fields with great achievements. In this paper, combining the on-line machine learning method into optimization algorithms is to be studied. In many heuristic optimization algorithms, one common way to reduce execution time and improve solution optimality is, first estimating the quality of a set of candidate solutions, and solving only promising candidates in detail. Currently most estimations are performed by empirical equations, whose accuracy greatly relies on the how well the equation is designed. In this paper, we propose an on-line learning based estimator to perform the solution estimation in heuristic algorithms to improve estimation accuracy. Then a simple case study is discussed, where a local search based heuristic with random start is used, and an on-line estimator considering the properties of local search is proposed. The experiments show that the accuracy of on-line estimator is much higher than the static estimator, and is also higher than a general off-line pre-Trained learner. Even though the on-line estimator introduced special time for its training, the heuristic algorithm still speeds up by 3.7X without optimality sacrifice.

AB - The study on machine learning has been flourishing for several years, and machine learning algorithms are being applied to various fields with great achievements. In this paper, combining the on-line machine learning method into optimization algorithms is to be studied. In many heuristic optimization algorithms, one common way to reduce execution time and improve solution optimality is, first estimating the quality of a set of candidate solutions, and solving only promising candidates in detail. Currently most estimations are performed by empirical equations, whose accuracy greatly relies on the how well the equation is designed. In this paper, we propose an on-line learning based estimator to perform the solution estimation in heuristic algorithms to improve estimation accuracy. Then a simple case study is discussed, where a local search based heuristic with random start is used, and an on-line estimator considering the properties of local search is proposed. The experiments show that the accuracy of on-line estimator is much higher than the static estimator, and is also higher than a general off-line pre-Trained learner. Even though the on-line estimator introduced special time for its training, the heuristic algorithm still speeds up by 3.7X without optimality sacrifice.

UR - http://www.scopus.com/inward/record.url?scp=85040790013&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85040790013&partnerID=8YFLogxK

U2 - 10.1109/CEEC.2017.8101593

DO - 10.1109/CEEC.2017.8101593

M3 - Conference contribution

AN - SCOPUS:85040790013

SP - 19

EP - 24

BT - 2017 9th Computer Science and Electronic Engineering Conference, CEEC 2017 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -