Online Diversity Control in Symbolic Regression via a Fast Hash-based Tree Similarity Measure

Bogdan Burlacu; Michael Affenzeller; Gabriel Kronberger; Michael Kommenda

doi:10.1109/CEC.2019.8790162

Online Diversity Control in Symbolic Regression via a Fast Hash-based Tree Similarity Measure

Research Center Hagenberg

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

11 Citations (Scopus)

Abstract

Diversity represents an important aspect of genetic programming, being directly correlated with search performance. When considered at the genotype level, diversity often requires expensive tree distance measures which have a negative impact on the algorithm’s runtime performance. In this work we introduce a fast, hash-based tree distance measure to massively speed-up the calculation of population diversity during the algorithmic run. We combine this measure with the standard GA and the NSGA-II genetic algorithms to steer the search towards higher diversity. We validate the approach on a collection of benchmark problems for symbolic regression where our method consistently outperforms the standard GA as well as NSGA-II configurations with different secondary objectives.

Original language	English
Title of host publication	2019 IEEE Congress on Evolutionary Computation, CEC 2019 - Proceedings
Publisher	IEEE Computational Intelligence Society
Pages	2175-2182
Number of pages	8
ISBN (Electronic)	9781728121536
ISBN (Print)	1089-778X
DOIs	https://doi.org/10.1109/CEC.2019.8790162
Publication status	Published - Jun 2019
Event	IEEE Congress on Evolutionary Computation - Wellington, New Zealand, Australia Duration: 10 Jun 2019 → 13 Jun 2019 http://cec2019.org/index.html

Publication series

Name	2019 IEEE Congress on Evolutionary Computation, CEC 2019 - Proceedings

Conference

Conference	IEEE Congress on Evolutionary Computation
Country/Territory	Australia
City	Wellington, New Zealand
Period	10.06.2019 → 13.06.2019
Internet address	http://cec2019.org/index.html

Keywords

symbolic regression
genetic programming
multi-objective symbolic regression
tree distance
genetic diversity
population diversity
tree hash
multi-objective

Access to Document

10.1109/CEC.2019.8790162

Cite this

Burlacu, B., Affenzeller, M., Kronberger, G., & Kommenda, M. (2019). Online Diversity Control in Symbolic Regression via a Fast Hash-based Tree Similarity Measure. In 2019 IEEE Congress on Evolutionary Computation, CEC 2019 - Proceedings (pp. 2175-2182). Article 8790162 (2019 IEEE Congress on Evolutionary Computation, CEC 2019 - Proceedings). IEEE Computational Intelligence Society. https://doi.org/10.1109/CEC.2019.8790162

Burlacu, Bogdan ; Affenzeller, Michael ; Kronberger, Gabriel et al. / Online Diversity Control in Symbolic Regression via a Fast Hash-based Tree Similarity Measure. 2019 IEEE Congress on Evolutionary Computation, CEC 2019 - Proceedings. IEEE Computational Intelligence Society, 2019. pp. 2175-2182 (2019 IEEE Congress on Evolutionary Computation, CEC 2019 - Proceedings).

@inproceedings{17f823488740434184013e70c1a1b743,

title = "Online Diversity Control in Symbolic Regression via a Fast Hash-based Tree Similarity Measure",

abstract = "Diversity represents an important aspect of genetic programming, being directly correlated with search performance. When considered at the genotype level, diversity often requires expensive tree distance measures which have a negative impact on the algorithm{\textquoteright}s runtime performance. In this work we introduce a fast, hash-based tree distance measure to massively speed-up the calculation of population diversity during the algorithmic run. We combine this measure with the standard GA and the NSGA-II genetic algorithms to steer the search towards higher diversity. We validate the approach on a collection of benchmark problems for symbolic regression where our method consistently outperforms the standard GA as well as NSGA-II configurations with different secondary objectives.",

keywords = "symbolic regression, genetic programming, multi-objective symbolic regression, tree distance, genetic diversity, symbolic regression, genetic programming, multi-objective symbolic regression, tree distance, genetic diversity, population diversity, tree hash, multi-objective",

author = "Bogdan Burlacu and Michael Affenzeller and Gabriel Kronberger and Michael Kommenda",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE. Copyright: Copyright 2020 Elsevier B.V., All rights reserved.; IEEE Congress on Evolutionary Computation ; Conference date: 10-06-2019 Through 13-06-2019",

year = "2019",

month = jun,

doi = "10.1109/CEC.2019.8790162",

language = "English",

isbn = "1089-778X",

series = "2019 IEEE Congress on Evolutionary Computation, CEC 2019 - Proceedings",

publisher = "IEEE Computational Intelligence Society",

pages = "2175--2182",

booktitle = "2019 IEEE Congress on Evolutionary Computation, CEC 2019 - Proceedings",

address = "United States",

url = "http://cec2019.org/index.html",

}

Burlacu, B , Affenzeller, M , Kronberger, G & Kommenda, M 2019, Online Diversity Control in Symbolic Regression via a Fast Hash-based Tree Similarity Measure. in 2019 IEEE Congress on Evolutionary Computation, CEC 2019 - Proceedings., 8790162, 2019 IEEE Congress on Evolutionary Computation, CEC 2019 - Proceedings, IEEE Computational Intelligence Society, pp. 2175-2182, IEEE Congress on Evolutionary Computation, Wellington, New Zealand, Australia, 10.06.2019. https://doi.org/10.1109/CEC.2019.8790162

Online Diversity Control in Symbolic Regression via a Fast Hash-based Tree Similarity Measure. / Burlacu, Bogdan ; Affenzeller, Michael ; Kronberger, Gabriel et al.
2019 IEEE Congress on Evolutionary Computation, CEC 2019 - Proceedings. IEEE Computational Intelligence Society, 2019. p. 2175-2182 8790162 (2019 IEEE Congress on Evolutionary Computation, CEC 2019 - Proceedings).

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

TY - GEN

T1 - Online Diversity Control in Symbolic Regression via a Fast Hash-based Tree Similarity Measure

AU - Burlacu, Bogdan

AU - Affenzeller, Michael

AU - Kronberger, Gabriel

AU - Kommenda, Michael

PY - 2019/6

Y1 - 2019/6

N2 - Diversity represents an important aspect of genetic programming, being directly correlated with search performance. When considered at the genotype level, diversity often requires expensive tree distance measures which have a negative impact on the algorithm’s runtime performance. In this work we introduce a fast, hash-based tree distance measure to massively speed-up the calculation of population diversity during the algorithmic run. We combine this measure with the standard GA and the NSGA-II genetic algorithms to steer the search towards higher diversity. We validate the approach on a collection of benchmark problems for symbolic regression where our method consistently outperforms the standard GA as well as NSGA-II configurations with different secondary objectives.

AB - Diversity represents an important aspect of genetic programming, being directly correlated with search performance. When considered at the genotype level, diversity often requires expensive tree distance measures which have a negative impact on the algorithm’s runtime performance. In this work we introduce a fast, hash-based tree distance measure to massively speed-up the calculation of population diversity during the algorithmic run. We combine this measure with the standard GA and the NSGA-II genetic algorithms to steer the search towards higher diversity. We validate the approach on a collection of benchmark problems for symbolic regression where our method consistently outperforms the standard GA as well as NSGA-II configurations with different secondary objectives.

KW - symbolic regression

KW - genetic programming

KW - multi-objective symbolic regression

KW - tree distance

KW - genetic diversity

KW - symbolic regression

KW - genetic programming

KW - multi-objective symbolic regression

KW - tree distance

KW - genetic diversity

KW - population diversity

KW - tree hash

KW - multi-objective

UR - https://www.scopus.com/pages/publications/85070592928

U2 - 10.1109/CEC.2019.8790162

DO - 10.1109/CEC.2019.8790162

M3 - Conference contribution

SN - 1089-778X

T3 - 2019 IEEE Congress on Evolutionary Computation, CEC 2019 - Proceedings

SP - 2175

EP - 2182

BT - 2019 IEEE Congress on Evolutionary Computation, CEC 2019 - Proceedings

PB - IEEE Computational Intelligence Society

T2 - IEEE Congress on Evolutionary Computation

Y2 - 10 June 2019 through 13 June 2019

ER -

Burlacu B , Affenzeller M , Kronberger G, Kommenda M. Online Diversity Control in Symbolic Regression via a Fast Hash-based Tree Similarity Measure. In 2019 IEEE Congress on Evolutionary Computation, CEC 2019 - Proceedings. IEEE Computational Intelligence Society. 2019. p. 2175-2182. 8790162. (2019 IEEE Congress on Evolutionary Computation, CEC 2019 - Proceedings). doi: 10.1109/CEC.2019.8790162

Online Diversity Control in Symbolic Regression via a Fast Hash-based Tree Similarity Measure

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this