On Switching Selection Methods to Increase Parsimony Pressure

Allan de Lima; Samuel Carvalho; Douglas Mota Dias; Joseph P. Sullivan; Conor Ryan

doi:10.5220/0012188900003595

On Switching Selection Methods to Increase Parsimony Pressure

Allan de Lima
, Samuel Carvalho
, Douglas Mota Dias
, Joseph P. Sullivan
, Conor Ryan

University of Limerick
Technological University of the Shannon: Midland Midwest

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

Abstract

We proposed a novel and simple selection system that alternates between tournament and Lexicase selection to tackle the bloat issue. In this way, we used Lexi², an implementation of Lexicase with lexicographic parsimony pressure, adopting the number of nodes in our solutions as size measurement. In addition, we increased the parsimony pressure by adding a penalty, also based on the number of nodes, to the aggregated fitness score. We analysed different scenarios, including some without extra parameters, in five benchmark problems: 2-bit Multiplier, 5-bit Parity, Car Evaluation, LED and Heart Disease. We succeeded in all of them in at least one scenario, reducing the size significantly while maintaining fitness. Beyond error and size, we also included results for the average number of fitness cases used in each generation.

Original language	English
Title of host publication	Proceedings of the 15th International Joint Conference on Computational Intelligence, IJCCI 2023
Editors	Niki van Stein, Francesco Marcelloni, H. K. Lam, Marie Cottrell, Joaquim Filipe
Publisher	Science and Technology Publications, Lda
Pages	96-107
Number of pages	12
ISBN (Electronic)	9789897586743
DOIs	https://doi.org/10.5220/0012188900003595
Publication status	Published - 2023
Event	15th International Joint Conference on Computational Intelligence, IJCCI 2023 - Hybrid, Rome, Italy Duration: 13 Nov 2023 → 15 Nov 2023

Publication series

Name	International Joint Conference on Computational Intelligence
ISSN (Electronic)	2184-3236

Conference

Conference	15th International Joint Conference on Computational Intelligence, IJCCI 2023
Country/Territory	Italy
City	Hybrid, Rome
Period	13/11/23 → 15/11/23

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Keywords

Bloat
Lexicase
Tournament

Access to Document

10.5220/0012188900003595

Cite this

de Lima, A., Carvalho, S., Dias, D. M., Sullivan, J. P., & Ryan, C. (2023). On Switching Selection Methods to Increase Parsimony Pressure. In N. van Stein, F. Marcelloni, H. K. Lam, M. Cottrell, & J. Filipe (Eds.), Proceedings of the 15th International Joint Conference on Computational Intelligence, IJCCI 2023 (pp. 96-107). (International Joint Conference on Computational Intelligence). Science and Technology Publications, Lda. https://doi.org/10.5220/0012188900003595

de Lima, Allan ; Carvalho, Samuel ; Dias, Douglas Mota et al. / On Switching Selection Methods to Increase Parsimony Pressure. Proceedings of the 15th International Joint Conference on Computational Intelligence, IJCCI 2023. editor / Niki van Stein ; Francesco Marcelloni ; H. K. Lam ; Marie Cottrell ; Joaquim Filipe. Science and Technology Publications, Lda, 2023. pp. 96-107 (International Joint Conference on Computational Intelligence).

@inproceedings{ed4835b5ef034fb2b587cad5e3f4b056,

title = "On Switching Selection Methods to Increase Parsimony Pressure",

abstract = "We proposed a novel and simple selection system that alternates between tournament and Lexicase selection to tackle the bloat issue. In this way, we used Lexi2, an implementation of Lexicase with lexicographic parsimony pressure, adopting the number of nodes in our solutions as size measurement. In addition, we increased the parsimony pressure by adding a penalty, also based on the number of nodes, to the aggregated fitness score. We analysed different scenarios, including some without extra parameters, in five benchmark problems: 2-bit Multiplier, 5-bit Parity, Car Evaluation, LED and Heart Disease. We succeeded in all of them in at least one scenario, reducing the size significantly while maintaining fitness. Beyond error and size, we also included results for the average number of fitness cases used in each generation.",

keywords = "Bloat, Lexicase, Tournament",

author = "\{de Lima\}, Allan and Samuel Carvalho and Dias, \{Douglas Mota\} and Sullivan, \{Joseph P.\} and Conor Ryan",

note = "Publisher Copyright: {\textcopyright} 2023 by SCITEPRESS – Science and Technology Publications, Lda.; 15th International Joint Conference on Computational Intelligence, IJCCI 2023 ; Conference date: 13-11-2023 Through 15-11-2023",

year = "2023",

doi = "10.5220/0012188900003595",

language = "English",

series = "International Joint Conference on Computational Intelligence",

publisher = "Science and Technology Publications, Lda",

pages = "96--107",

editor = "\{van Stein\}, Niki and Francesco Marcelloni and Lam, \{H. K.\} and Marie Cottrell and Joaquim Filipe",

booktitle = "Proceedings of the 15th International Joint Conference on Computational Intelligence, IJCCI 2023",

}

de Lima, A, Carvalho, S, Dias, DM, Sullivan, JP & Ryan, C 2023, On Switching Selection Methods to Increase Parsimony Pressure. in N van Stein, F Marcelloni, HK Lam, M Cottrell & J Filipe (eds), Proceedings of the 15th International Joint Conference on Computational Intelligence, IJCCI 2023. International Joint Conference on Computational Intelligence, Science and Technology Publications, Lda, pp. 96-107, 15th International Joint Conference on Computational Intelligence, IJCCI 2023, Hybrid, Rome, Italy, 13/11/23. https://doi.org/10.5220/0012188900003595

On Switching Selection Methods to Increase Parsimony Pressure. / de Lima, Allan; Carvalho, Samuel; Dias, Douglas Mota et al.
Proceedings of the 15th International Joint Conference on Computational Intelligence, IJCCI 2023. ed. / Niki van Stein; Francesco Marcelloni; H. K. Lam; Marie Cottrell; Joaquim Filipe. Science and Technology Publications, Lda, 2023. p. 96-107 (International Joint Conference on Computational Intelligence).

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

TY - GEN

T1 - On Switching Selection Methods to Increase Parsimony Pressure

AU - de Lima, Allan

AU - Carvalho, Samuel

AU - Dias, Douglas Mota

AU - Sullivan, Joseph P.

AU - Ryan, Conor

PY - 2023

Y1 - 2023

N2 - We proposed a novel and simple selection system that alternates between tournament and Lexicase selection to tackle the bloat issue. In this way, we used Lexi2, an implementation of Lexicase with lexicographic parsimony pressure, adopting the number of nodes in our solutions as size measurement. In addition, we increased the parsimony pressure by adding a penalty, also based on the number of nodes, to the aggregated fitness score. We analysed different scenarios, including some without extra parameters, in five benchmark problems: 2-bit Multiplier, 5-bit Parity, Car Evaluation, LED and Heart Disease. We succeeded in all of them in at least one scenario, reducing the size significantly while maintaining fitness. Beyond error and size, we also included results for the average number of fitness cases used in each generation.

AB - We proposed a novel and simple selection system that alternates between tournament and Lexicase selection to tackle the bloat issue. In this way, we used Lexi2, an implementation of Lexicase with lexicographic parsimony pressure, adopting the number of nodes in our solutions as size measurement. In addition, we increased the parsimony pressure by adding a penalty, also based on the number of nodes, to the aggregated fitness score. We analysed different scenarios, including some without extra parameters, in five benchmark problems: 2-bit Multiplier, 5-bit Parity, Car Evaluation, LED and Heart Disease. We succeeded in all of them in at least one scenario, reducing the size significantly while maintaining fitness. Beyond error and size, we also included results for the average number of fitness cases used in each generation.

KW - Bloat

KW - Lexicase

KW - Tournament

UR - https://www.scopus.com/pages/publications/85188240424

U2 - 10.5220/0012188900003595

DO - 10.5220/0012188900003595

M3 - Conference contribution

AN - SCOPUS:85188240424

T3 - International Joint Conference on Computational Intelligence

SP - 96

EP - 107

BT - Proceedings of the 15th International Joint Conference on Computational Intelligence, IJCCI 2023

A2 - van Stein, Niki

A2 - Marcelloni, Francesco

A2 - Lam, H. K.

A2 - Cottrell, Marie

A2 - Filipe, Joaquim

PB - Science and Technology Publications, Lda

T2 - 15th International Joint Conference on Computational Intelligence, IJCCI 2023

Y2 - 13 November 2023 through 15 November 2023

ER -

de Lima A, Carvalho S, Dias DM, Sullivan JP, Ryan C. On Switching Selection Methods to Increase Parsimony Pressure. In van Stein N, Marcelloni F, Lam HK, Cottrell M, Filipe J, editors, Proceedings of the 15th International Joint Conference on Computational Intelligence, IJCCI 2023. Science and Technology Publications, Lda. 2023. p. 96-107. (International Joint Conference on Computational Intelligence). doi: 10.5220/0012188900003595

On Switching Selection Methods to Increase Parsimony Pressure

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this