Using gradient-based information to deal with scalability in multi-objective evolutionary algorithms

Adriana Lara; Carlos A. Coello Coello; Oliver Schütze

doi:10.1109/CEC.2009.4982925

Using gradient-based information to deal with scalability in multi-objective evolutionary algorithms

Adriana Lara, Carlos A. Coello Coello, Oliver Schütze

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

9 Scopus citations

Abstract

This work introduces a hybrid between an elitist multi-objective evolutionary algorithm and a gradient-based descent method, which is applied only to certain (selected) solutions. Our proposed approach requires a low number of objective function evaluations to converge to a few points in the Pareto front. Then, the rest of the Pareto front is reconstructed using a method based on rough sets theory, which also requires a low number of objective function evaluations. Emphasis is placed on the effectiveness of our proposed hybrid approach when increasing the number of decision variables, and a study of the scalability of our approach is also presented.

Original language	English
Title of host publication	2009 IEEE Congress on Evolutionary Computation, CEC 2009
Pages	16-23
Number of pages	8
DOIs	https://doi.org/10.1109/CEC.2009.4982925
State	Published - 2009
Externally published	Yes
Event	2009 IEEE Congress on Evolutionary Computation, CEC 2009 - Trondheim, Norway Duration: 18 May 2009 → 21 May 2009

Publication series

Name	2009 IEEE Congress on Evolutionary Computation, CEC 2009

Conference

Conference	2009 IEEE Congress on Evolutionary Computation, CEC 2009
Country/Territory	Norway
City	Trondheim
Period	18/05/09 → 21/05/09

Access to Document

10.1109/CEC.2009.4982925

Cite this

@inproceedings{ee0032c821f7422da244c0ff4ba059f8,

title = "Using gradient-based information to deal with scalability in multi-objective evolutionary algorithms",

abstract = "This work introduces a hybrid between an elitist multi-objective evolutionary algorithm and a gradient-based descent method, which is applied only to certain (selected) solutions. Our proposed approach requires a low number of objective function evaluations to converge to a few points in the Pareto front. Then, the rest of the Pareto front is reconstructed using a method based on rough sets theory, which also requires a low number of objective function evaluations. Emphasis is placed on the effectiveness of our proposed hybrid approach when increasing the number of decision variables, and a study of the scalability of our approach is also presented.",

author = "Adriana Lara and {Coello Coello}, {Carlos A.} and Oliver Sch{\"u}tze",

year = "2009",

doi = "10.1109/CEC.2009.4982925",

language = "Ingl{\'e}s",

isbn = "9781424429592",

series = "2009 IEEE Congress on Evolutionary Computation, CEC 2009",

pages = "16--23",

booktitle = "2009 IEEE Congress on Evolutionary Computation, CEC 2009",

note = "2009 IEEE Congress on Evolutionary Computation, CEC 2009 ; Conference date: 18-05-2009 Through 21-05-2009",

}

Lara, A, Coello Coello, CA & Schütze, O 2009, Using gradient-based information to deal with scalability in multi-objective evolutionary algorithms. in 2009 IEEE Congress on Evolutionary Computation, CEC 2009., 4982925, 2009 IEEE Congress on Evolutionary Computation, CEC 2009, pp. 16-23, 2009 IEEE Congress on Evolutionary Computation, CEC 2009, Trondheim, Norway, 18/05/09. https://doi.org/10.1109/CEC.2009.4982925

Using gradient-based information to deal with scalability in multi-objective evolutionary algorithms. / Lara, Adriana; Coello Coello, Carlos A.; Schütze, Oliver.
2009 IEEE Congress on Evolutionary Computation, CEC 2009. 2009. p. 16-23 4982925 (2009 IEEE Congress on Evolutionary Computation, CEC 2009).

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

TY - GEN

T1 - Using gradient-based information to deal with scalability in multi-objective evolutionary algorithms

AU - Lara, Adriana

AU - Coello Coello, Carlos A.

AU - Schütze, Oliver

PY - 2009

Y1 - 2009

N2 - This work introduces a hybrid between an elitist multi-objective evolutionary algorithm and a gradient-based descent method, which is applied only to certain (selected) solutions. Our proposed approach requires a low number of objective function evaluations to converge to a few points in the Pareto front. Then, the rest of the Pareto front is reconstructed using a method based on rough sets theory, which also requires a low number of objective function evaluations. Emphasis is placed on the effectiveness of our proposed hybrid approach when increasing the number of decision variables, and a study of the scalability of our approach is also presented.

AB - This work introduces a hybrid between an elitist multi-objective evolutionary algorithm and a gradient-based descent method, which is applied only to certain (selected) solutions. Our proposed approach requires a low number of objective function evaluations to converge to a few points in the Pareto front. Then, the rest of the Pareto front is reconstructed using a method based on rough sets theory, which also requires a low number of objective function evaluations. Emphasis is placed on the effectiveness of our proposed hybrid approach when increasing the number of decision variables, and a study of the scalability of our approach is also presented.

UR - http://www.scopus.com/inward/record.url?scp=70449817444&partnerID=8YFLogxK

U2 - 10.1109/CEC.2009.4982925

DO - 10.1109/CEC.2009.4982925

M3 - Contribución a la conferencia

AN - SCOPUS:70449817444

SN - 9781424429592

T3 - 2009 IEEE Congress on Evolutionary Computation, CEC 2009

SP - 16

EP - 23

BT - 2009 IEEE Congress on Evolutionary Computation, CEC 2009

T2 - 2009 IEEE Congress on Evolutionary Computation, CEC 2009

Y2 - 18 May 2009 through 21 May 2009

ER -

Using gradient-based information to deal with scalability in multi-objective evolutionary algorithms

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this