On the Performance of Master-Slave Parallelization Methods for Multi-Objective Evolutionary Algorithms

Ciprian Zavoianu; Edwin Lughofer; Werner Koppelstätter; Günther Weidenholzer; Wolfgang Amrhein; Erich Klement

doi:10.1007/978-3-642-38610-7_12

On the Performance of Master-Slave Parallelization Methods for Multi-Objective Evolutionary Algorithms

Ciprian Zavoianu
, Edwin Lughofer
, Werner Koppelstätter
, Günther Weidenholzer
, Wolfgang Amrhein
, Erich Klement

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

Abstract

This paper is focused on a comparative analysis of the performance of two master-slave parallelization methods, the basic generational scheme and the steady-state asynchronous scheme. Both can be used to improve the convergence speed of multi-objective evolutionary algorithms (MOEAs) that rely on time-intensive fitness evaluation functions. The importance of this work stems from the fact that a correct choice for one or the other parallelization method can lead to considerable speed improvements with regards to the overall duration of the optimization. Our main aim is to provide practitioners of MOEAs with a simple but effective method of deciding which master-slave parallelization option is better when dealing with a time-constrained optimization process.

Original language	English
Title of host publication	Artificial Intelligence and Soft Computing
Editors	Laszek Rutkowski and Marcin Korytkowski and Rafal Scherer and Ryszard Tadeusiewicz and Lotfi A. Zadeh and Jacek M. Zurada
Publisher	Springer Berlin Heidelberg
Pages	122-134
Number of pages	13
Volume	7895
ISBN (Print)	978-3-642-38609-1
DOIs	https://doi.org/10.1007/978-3-642-38610-7_12
Publication status	Published - Jun 2013

Publication series

Name	Lecture Notes in Artificial Intelligence (LNAI)

Fields of science

101001 Algebra
101 Mathematics
102 Computer Sciences
101013 Mathematical logic
101020 Technical mathematics
102001 Artificial intelligence
102003 Image processing
202027 Mechatronics
101019 Stochastics
211913 Quality assurance

JKU Focus areas

Computation in Informatics and Mathematics
Mechatronics and Information Processing

Access to Document

10.1007/978-3-642-38610-7_12

Cite this

Zavoianu, C., Lughofer, E., Koppelstätter, W., Weidenholzer, G., Amrhein, W., & Klement, E. (2013). On the Performance of Master-Slave Parallelization Methods for Multi-Objective Evolutionary Algorithms. In Laszek Rutkowski and Marcin Korytkowski and Rafal Scherer and Ryszard Tadeusiewicz and Lotfi A. Zadeh and Jacek M. Zurada (Ed.), Artificial Intelligence and Soft Computing (Vol. 7895, pp. 122-134). (Lecture Notes in Artificial Intelligence (LNAI)). Springer Berlin Heidelberg. https://doi.org/10.1007/978-3-642-38610-7_12

Zavoianu, Ciprian ; Lughofer, Edwin ; Koppelstätter, Werner et al. / On the Performance of Master-Slave Parallelization Methods for Multi-Objective Evolutionary Algorithms. Artificial Intelligence and Soft Computing. editor / Laszek Rutkowski and Marcin Korytkowski and Rafal Scherer and Ryszard Tadeusiewicz and Lotfi A. Zadeh and Jacek M. Zurada. Vol. 7895 Springer Berlin Heidelberg, 2013. pp. 122-134 (Lecture Notes in Artificial Intelligence (LNAI)).

@inproceedings{34a50181d2614fa4bdc208f8808ee771,

title = "On the Performance of Master-Slave Parallelization Methods for Multi-Objective Evolutionary Algorithms",

abstract = "This paper is focused on a comparative analysis of the performance of two master-slave parallelization methods, the basic generational scheme and the steady-state asynchronous scheme. Both can be used to improve the convergence speed of multi-objective evolutionary algorithms (MOEAs) that rely on time-intensive fitness evaluation functions. The importance of this work stems from the fact that a correct choice for one or the other parallelization method can lead to considerable speed improvements with regards to the overall duration of the optimization. Our main aim is to provide practitioners of MOEAs with a simple but effective method of deciding which master-slave parallelization option is better when dealing with a time-constrained optimization process.",

keywords = "evolutionary computation, master-slave parallelization, multi-objective optimization, performance comparison, steady-state evolution",

author = "Ciprian Zavoianu and Edwin Lughofer and Werner Koppelst{\"a}tter and G{\"u}nther Weidenholzer and Wolfgang Amrhein and Erich Klement",

year = "2013",

month = jun,

doi = "10.1007/978-3-642-38610-7\_12",

language = "English",

isbn = "978-3-642-38609-1",

volume = "7895",

series = "Lecture Notes in Artificial Intelligence (LNAI)",

publisher = "Springer Berlin Heidelberg",

pages = "122--134",

editor = "\{Laszek Rutkowski and Marcin Korytkowski and Rafal Scherer and Ryszard Tadeusiewicz and Lotfi A. Zadeh and Jacek M. Zurada\}",

booktitle = "Artificial Intelligence and Soft Computing",

}

Zavoianu, C, Lughofer, E, Koppelstätter, W, Weidenholzer, G, Amrhein, W & Klement, E 2013, On the Performance of Master-Slave Parallelization Methods for Multi-Objective Evolutionary Algorithms. in Laszek Rutkowski and Marcin Korytkowski and Rafal Scherer and Ryszard Tadeusiewicz and Lotfi A. Zadeh and Jacek M. Zurada (ed.), Artificial Intelligence and Soft Computing. vol. 7895, Lecture Notes in Artificial Intelligence (LNAI), Springer Berlin Heidelberg, pp. 122-134. https://doi.org/10.1007/978-3-642-38610-7_12

On the Performance of Master-Slave Parallelization Methods for Multi-Objective Evolutionary Algorithms. / Zavoianu, Ciprian; Lughofer, Edwin; Koppelstätter, Werner et al.
Artificial Intelligence and Soft Computing. ed. / Laszek Rutkowski and Marcin Korytkowski and Rafal Scherer and Ryszard Tadeusiewicz and Lotfi A. Zadeh and Jacek M. Zurada. Vol. 7895 Springer Berlin Heidelberg, 2013. p. 122-134 (Lecture Notes in Artificial Intelligence (LNAI)).

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

TY - GEN

T1 - On the Performance of Master-Slave Parallelization Methods for Multi-Objective Evolutionary Algorithms

AU - Zavoianu, Ciprian

AU - Lughofer, Edwin

AU - Koppelstätter, Werner

AU - Weidenholzer, Günther

AU - Amrhein, Wolfgang

AU - Klement, Erich

PY - 2013/6

Y1 - 2013/6

N2 - This paper is focused on a comparative analysis of the performance of two master-slave parallelization methods, the basic generational scheme and the steady-state asynchronous scheme. Both can be used to improve the convergence speed of multi-objective evolutionary algorithms (MOEAs) that rely on time-intensive fitness evaluation functions. The importance of this work stems from the fact that a correct choice for one or the other parallelization method can lead to considerable speed improvements with regards to the overall duration of the optimization. Our main aim is to provide practitioners of MOEAs with a simple but effective method of deciding which master-slave parallelization option is better when dealing with a time-constrained optimization process.

AB - This paper is focused on a comparative analysis of the performance of two master-slave parallelization methods, the basic generational scheme and the steady-state asynchronous scheme. Both can be used to improve the convergence speed of multi-objective evolutionary algorithms (MOEAs) that rely on time-intensive fitness evaluation functions. The importance of this work stems from the fact that a correct choice for one or the other parallelization method can lead to considerable speed improvements with regards to the overall duration of the optimization. Our main aim is to provide practitioners of MOEAs with a simple but effective method of deciding which master-slave parallelization option is better when dealing with a time-constrained optimization process.

KW - evolutionary computation

KW - master-slave parallelization

KW - multi-objective optimization

KW - performance comparison

KW - steady-state evolution

UR - https://www.scopus.com/pages/publications/84884401293

U2 - 10.1007/978-3-642-38610-7_12

DO - 10.1007/978-3-642-38610-7_12

M3 - Conference proceedings

SN - 978-3-642-38609-1

VL - 7895

T3 - Lecture Notes in Artificial Intelligence (LNAI)

SP - 122

EP - 134

BT - Artificial Intelligence and Soft Computing

A2 - Laszek Rutkowski and Marcin Korytkowski and Rafal Scherer and Ryszard Tadeusiewicz and Lotfi A. Zadeh and Jacek M. Zurada, null

PB - Springer Berlin Heidelberg

ER -

Zavoianu C, Lughofer E, Koppelstätter W, Weidenholzer G, Amrhein W , Klement E. On the Performance of Master-Slave Parallelization Methods for Multi-Objective Evolutionary Algorithms. In Laszek Rutkowski and Marcin Korytkowski and Rafal Scherer and Ryszard Tadeusiewicz and Lotfi A. Zadeh and Jacek M. Zurada, editor, Artificial Intelligence and Soft Computing. Vol. 7895. Springer Berlin Heidelberg. 2013. p. 122-134. (Lecture Notes in Artificial Intelligence (LNAI)). doi: 10.1007/978-3-642-38610-7_12

On the Performance of Master-Slave Parallelization Methods for Multi-Objective Evolutionary Algorithms

Abstract

Publication series

Fields of science

JKU Focus areas

Access to Document

Other files and links

Cite this