Alignment-based genetic programming for real life applications

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

A recent discovery has attracted the attention of many researchers in the field of genetic programming: given individuals with particular characteristics of alignment in the error space, called optimally aligned, it is possible to reconstruct a globally optimal solution. Furthermore, recent preliminary experiments have shown that an indirect search consisting of looking for optimally aligned individuals can have benefits in terms of generalization ability compared to a direct search for optimal solutions. For this reason, defining genetic programming systems that look for optimally aligned individuals is becoming an ambitious and important objective. Nevertheless, the systems that have been introduced so far present important limitations that make them unusable in practice, particularly for complex real-life applications. In this paper, we overcome those limitations, and we present the first usable alignment-based genetic programming system, called nested alignment genetic programming (NAGP). The presented experimental results show that NAGP is able to outperform two of the most recognized state-of-the-art genetic programming systems on four complex real-life applications. The predictive models generated by NAGP are not only more effective than the ones produced by the other studied methods but also significantly smaller and thus more manageable and interpretable.

Original languageEnglish
Pages (from-to)840-851
Number of pages12
JournalSwarm and Evolutionary Computation
Volume44
Issue numberFebruary
Early online date29 Sep 2018
DOIs
Publication statusPublished - Feb 2019

Fingerprint

Genetic programming
Genetic Programming
Alignment
Optimal Solution
Direct Search
Predictive Model
Life
Experimental Results
Experiment

Keywords

  • Alignment
  • Error space
  • Genetic programming
  • Geometric semantic operators
  • Real-life applications

Cite this

@article{c56bbced5fe94ef3a57a40ae13fa0353,
title = "Alignment-based genetic programming for real life applications",
abstract = "A recent discovery has attracted the attention of many researchers in the field of genetic programming: given individuals with particular characteristics of alignment in the error space, called optimally aligned, it is possible to reconstruct a globally optimal solution. Furthermore, recent preliminary experiments have shown that an indirect search consisting of looking for optimally aligned individuals can have benefits in terms of generalization ability compared to a direct search for optimal solutions. For this reason, defining genetic programming systems that look for optimally aligned individuals is becoming an ambitious and important objective. Nevertheless, the systems that have been introduced so far present important limitations that make them unusable in practice, particularly for complex real-life applications. In this paper, we overcome those limitations, and we present the first usable alignment-based genetic programming system, called nested alignment genetic programming (NAGP). The presented experimental results show that NAGP is able to outperform two of the most recognized state-of-the-art genetic programming systems on four complex real-life applications. The predictive models generated by NAGP are not only more effective than the ones produced by the other studied methods but also significantly smaller and thus more manageable and interpretable.",
keywords = "Alignment, Error space, Genetic programming, Geometric semantic operators, Real-life applications",
author = "Leonardo Vanneschi and Mauro Castelli and Kristen Scott and Leonardo Trujillo",
note = "Vanneschi, L., Castelli, M., Scott, K., & Trujillo, L. (2019). Alignment-based genetic programming for real life applications. Swarm and Evolutionary Computation, 44(February), 840-851. DOI: 10.1016/j.swevo.2018.09.006",
year = "2019",
month = "2",
doi = "10.1016/j.swevo.2018.09.006",
language = "English",
volume = "44",
pages = "840--851",
journal = "Swarm and Evolutionary Computation",
issn = "2210-6502",
publisher = "Elsevier",
number = "February",

}

Alignment-based genetic programming for real life applications. / Vanneschi, Leonardo; Castelli, Mauro; Scott, Kristen; Trujillo, Leonardo.

In: Swarm and Evolutionary Computation, Vol. 44, No. February, 02.2019, p. 840-851.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Alignment-based genetic programming for real life applications

AU - Vanneschi, Leonardo

AU - Castelli, Mauro

AU - Scott, Kristen

AU - Trujillo, Leonardo

N1 - Vanneschi, L., Castelli, M., Scott, K., & Trujillo, L. (2019). Alignment-based genetic programming for real life applications. Swarm and Evolutionary Computation, 44(February), 840-851. DOI: 10.1016/j.swevo.2018.09.006

PY - 2019/2

Y1 - 2019/2

N2 - A recent discovery has attracted the attention of many researchers in the field of genetic programming: given individuals with particular characteristics of alignment in the error space, called optimally aligned, it is possible to reconstruct a globally optimal solution. Furthermore, recent preliminary experiments have shown that an indirect search consisting of looking for optimally aligned individuals can have benefits in terms of generalization ability compared to a direct search for optimal solutions. For this reason, defining genetic programming systems that look for optimally aligned individuals is becoming an ambitious and important objective. Nevertheless, the systems that have been introduced so far present important limitations that make them unusable in practice, particularly for complex real-life applications. In this paper, we overcome those limitations, and we present the first usable alignment-based genetic programming system, called nested alignment genetic programming (NAGP). The presented experimental results show that NAGP is able to outperform two of the most recognized state-of-the-art genetic programming systems on four complex real-life applications. The predictive models generated by NAGP are not only more effective than the ones produced by the other studied methods but also significantly smaller and thus more manageable and interpretable.

AB - A recent discovery has attracted the attention of many researchers in the field of genetic programming: given individuals with particular characteristics of alignment in the error space, called optimally aligned, it is possible to reconstruct a globally optimal solution. Furthermore, recent preliminary experiments have shown that an indirect search consisting of looking for optimally aligned individuals can have benefits in terms of generalization ability compared to a direct search for optimal solutions. For this reason, defining genetic programming systems that look for optimally aligned individuals is becoming an ambitious and important objective. Nevertheless, the systems that have been introduced so far present important limitations that make them unusable in practice, particularly for complex real-life applications. In this paper, we overcome those limitations, and we present the first usable alignment-based genetic programming system, called nested alignment genetic programming (NAGP). The presented experimental results show that NAGP is able to outperform two of the most recognized state-of-the-art genetic programming systems on four complex real-life applications. The predictive models generated by NAGP are not only more effective than the ones produced by the other studied methods but also significantly smaller and thus more manageable and interpretable.

KW - Alignment

KW - Error space

KW - Genetic programming

KW - Geometric semantic operators

KW - Real-life applications

UR - http://www.scopus.com/inward/record.url?scp=85054516429&partnerID=8YFLogxK

UR - http://gateway.webofknowledge.com/gateway/Gateway.cgi?GWVersion=2&SrcAuth=Alerting&SrcApp=Alerting&DestApp=WOS_CPL&DestLinkType=FullRecord&UT=WOS:000456761600058

U2 - 10.1016/j.swevo.2018.09.006

DO - 10.1016/j.swevo.2018.09.006

M3 - Article

VL - 44

SP - 840

EP - 851

JO - Swarm and Evolutionary Computation

JF - Swarm and Evolutionary Computation

SN - 2210-6502

IS - February

ER -