Enhancing the performance of GP using an ancestry-based mate selection scheme

R Fry; A Tyrrell

Enhancing the performance of GP using an ancestry-based mate selection scheme

Electronic Engineering

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

Abstract

The performance of genetic programming relies mostly on population-contained variation. If the population diversity is low then there will be a greater chance of the algorithm being unable to find the global optimum. We present a new method of approximating the genetic similarity between two individuals using ancestry information. We define a new diversity-preserving selection scheme, based on standard tournament selection, which encourages genetically dissimilar individuals to undergo genetic operation. The new method is illustrated by assessing its performance in a well-known problem domain: algebraic symbolic regression.

Original language	English
Title of host publication	GENETIC AND EVOLUTIONARY COMPUTATION - GECCO 2003, PT II, PROCEEDINGS
Editors	E CantuPaz, JA Foster, K Deb, LD Davis, R Roy, UM OReilly, HG Beyer, R Standish, G Kendall, S Wilson, M Hartman, J Wegener, D Dasgupta, MA Potter, AC Schultz, KA Dowsland, N Jonoska, J Miller
Place of Publication	BERLIN
Publisher	Springer
Pages	1804-1805
Number of pages	2
ISBN (Print)	3-540-40603-4
Publication status	Published - 2003
Event	5th Annual Genetic and Evolutionary Computation Conference (GECCO 2003) - CHICAGO Duration: 12 Jul 2003 → 16 Jul 2003

Conference

Conference	5th Annual Genetic and Evolutionary Computation Conference (GECCO 2003)
City	CHICAGO
Period	12/07/03 → 16/07/03

Cite this

Fry, R., & Tyrrell, A. (2003). Enhancing the performance of GP using an ancestry-based mate selection scheme. In E. CantuPaz, JA. Foster, K. Deb, LD. Davis, R. Roy, UM. OReilly, HG. Beyer, R. Standish, G. Kendall, S. Wilson, M. Hartman, J. Wegener, D. Dasgupta, MA. Potter, AC. Schultz, KA. Dowsland, N. Jonoska, & J. Miller (Eds.), GENETIC AND EVOLUTIONARY COMPUTATION - GECCO 2003, PT II, PROCEEDINGS (pp. 1804-1805). Springer.

Fry, R ; Tyrrell, A. / Enhancing the performance of GP using an ancestry-based mate selection scheme. GENETIC AND EVOLUTIONARY COMPUTATION - GECCO 2003, PT II, PROCEEDINGS. editor / E CantuPaz ; JA Foster ; K Deb ; LD Davis ; R Roy ; UM OReilly ; HG Beyer ; R Standish ; G Kendall ; S Wilson ; M Hartman ; J Wegener ; D Dasgupta ; MA Potter ; AC Schultz ; KA Dowsland ; N Jonoska ; J Miller. BERLIN : Springer, 2003. pp. 1804-1805

@inproceedings{e549aca28b074d5ea64acccd5dafb8c6,

title = "Enhancing the performance of GP using an ancestry-based mate selection scheme",

abstract = "The performance of genetic programming relies mostly on population-contained variation. If the population diversity is low then there will be a greater chance of the algorithm being unable to find the global optimum. We present a new method of approximating the genetic similarity between two individuals using ancestry information. We define a new diversity-preserving selection scheme, based on standard tournament selection, which encourages genetically dissimilar individuals to undergo genetic operation. The new method is illustrated by assessing its performance in a well-known problem domain: algebraic symbolic regression.",

author = "R Fry and A Tyrrell",

year = "2003",

language = "English",

isbn = "3-540-40603-4",

pages = "1804--1805",

editor = "E CantuPaz and JA Foster and K Deb and LD Davis and R Roy and UM OReilly and HG Beyer and R Standish and G Kendall and S Wilson and M Hartman and J Wegener and D Dasgupta and MA Potter and AC Schultz and KA Dowsland and N Jonoska and J Miller",

booktitle = "GENETIC AND EVOLUTIONARY COMPUTATION - GECCO 2003, PT II, PROCEEDINGS",

publisher = "Springer",

address = "Germany",

note = "5th Annual Genetic and Evolutionary Computation Conference (GECCO 2003) ; Conference date: 12-07-2003 Through 16-07-2003",

}

Fry, R & Tyrrell, A 2003, Enhancing the performance of GP using an ancestry-based mate selection scheme. in E CantuPaz, JA Foster, K Deb, LD Davis, R Roy, UM OReilly, HG Beyer, R Standish, G Kendall, S Wilson, M Hartman, J Wegener, D Dasgupta, MA Potter, AC Schultz, KA Dowsland, N Jonoska & J Miller (eds), GENETIC AND EVOLUTIONARY COMPUTATION - GECCO 2003, PT II, PROCEEDINGS. Springer, BERLIN, pp. 1804-1805, 5th Annual Genetic and Evolutionary Computation Conference (GECCO 2003), CHICAGO, 12/07/03.

Enhancing the performance of GP using an ancestry-based mate selection scheme. / Fry, R; Tyrrell, A.

GENETIC AND EVOLUTIONARY COMPUTATION - GECCO 2003, PT II, PROCEEDINGS. ed. / E CantuPaz; JA Foster; K Deb; LD Davis; R Roy; UM OReilly; HG Beyer; R Standish; G Kendall; S Wilson; M Hartman; J Wegener; D Dasgupta; MA Potter; AC Schultz; KA Dowsland; N Jonoska; J Miller. BERLIN : Springer, 2003. p. 1804-1805.

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

TY - GEN

T1 - Enhancing the performance of GP using an ancestry-based mate selection scheme

AU - Fry, R

AU - Tyrrell, A

PY - 2003

Y1 - 2003

N2 - The performance of genetic programming relies mostly on population-contained variation. If the population diversity is low then there will be a greater chance of the algorithm being unable to find the global optimum. We present a new method of approximating the genetic similarity between two individuals using ancestry information. We define a new diversity-preserving selection scheme, based on standard tournament selection, which encourages genetically dissimilar individuals to undergo genetic operation. The new method is illustrated by assessing its performance in a well-known problem domain: algebraic symbolic regression.

AB - The performance of genetic programming relies mostly on population-contained variation. If the population diversity is low then there will be a greater chance of the algorithm being unable to find the global optimum. We present a new method of approximating the genetic similarity between two individuals using ancestry information. We define a new diversity-preserving selection scheme, based on standard tournament selection, which encourages genetically dissimilar individuals to undergo genetic operation. The new method is illustrated by assessing its performance in a well-known problem domain: algebraic symbolic regression.

M3 - Conference contribution

SN - 3-540-40603-4

SP - 1804

EP - 1805

BT - GENETIC AND EVOLUTIONARY COMPUTATION - GECCO 2003, PT II, PROCEEDINGS

A2 - CantuPaz, E

A2 - Foster, JA

A2 - Deb, K

A2 - Davis, LD

A2 - Roy, R

A2 - OReilly, UM

A2 - Beyer, HG

A2 - Standish, R

A2 - Kendall, G

A2 - Wilson, S

A2 - Hartman, M

A2 - Wegener, J

A2 - Dasgupta, D

A2 - Potter, MA

A2 - Schultz, AC

A2 - Dowsland, KA

A2 - Jonoska, N

A2 - Miller, J

PB - Springer

CY - BERLIN

T2 - 5th Annual Genetic and Evolutionary Computation Conference (GECCO 2003)

Y2 - 12 July 2003 through 16 July 2003

ER -

Fry R, Tyrrell A. Enhancing the performance of GP using an ancestry-based mate selection scheme. In CantuPaz E, Foster JA, Deb K, Davis LD, Roy R, OReilly UM, Beyer HG, Standish R, Kendall G, Wilson S, Hartman M, Wegener J, Dasgupta D, Potter MA, Schultz AC, Dowsland KA, Jonoska N, Miller J, editors, GENETIC AND EVOLUTIONARY COMPUTATION - GECCO 2003, PT II, PROCEEDINGS. BERLIN: Springer. 2003. p. 1804-1805