Extended rule-based genetic network programming

Xianneng Li, Kotaro Hirasawa

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Citations (Scopus)

Abstract

Recent advances in rule-based systems, i.e., Learning Classifier Systems (LCSs), have shown their sequential decisionmaking ability with a generalization property. In this paper, a novel LCS named eXtended rule-based Genetic Network Programming (XrGNP) is proposed. Different from most of the current LCSs, the rules are represented and discovered through a graph-based evolutionary algorithm GNP, which consequently has the distinct expression ability to model and evolve the "if-then" decision-making rules. Experiments on a benchmark multi-step problem (so-called Reinforcement Learning problem) demonstrate its effectiveness.

Original languageEnglish
Title of host publicationGECCO 2013 - Proceedings of the 2013 Genetic and Evolutionary Computation Conference Companion
Pages155-156
Number of pages2
DOIs
Publication statusPublished - 2013
Event15th Annual Conference on Genetic and Evolutionary Computation, GECCO 2013 - Amsterdam
Duration: 2013 Jul 62013 Jul 10

Other

Other15th Annual Conference on Genetic and Evolutionary Computation, GECCO 2013
CityAmsterdam
Period13/7/613/7/10

Fingerprint

Network Programming
Learning Classifier Systems
Genetic Network
Genetic Programming
Classifiers
Decision Making
Rule-based Systems
Reinforcement learning
Knowledge based systems
Reinforcement Learning
Evolutionary algorithms
Evolutionary Algorithms
Decision making
Benchmark
Distinct
Graph in graph theory
Demonstrate
Experiment
Experiments
Model

Keywords

  • Genetic network programming
  • Learning classifier systems
  • XrGNP

ASJC Scopus subject areas

  • Computational Mathematics

Cite this

Li, X., & Hirasawa, K. (2013). Extended rule-based genetic network programming. In GECCO 2013 - Proceedings of the 2013 Genetic and Evolutionary Computation Conference Companion (pp. 155-156) https://doi.org/10.1145/2464576.2464655

Extended rule-based genetic network programming. / Li, Xianneng; Hirasawa, Kotaro.

GECCO 2013 - Proceedings of the 2013 Genetic and Evolutionary Computation Conference Companion. 2013. p. 155-156.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Li, X & Hirasawa, K 2013, Extended rule-based genetic network programming. in GECCO 2013 - Proceedings of the 2013 Genetic and Evolutionary Computation Conference Companion. pp. 155-156, 15th Annual Conference on Genetic and Evolutionary Computation, GECCO 2013, Amsterdam, 13/7/6. https://doi.org/10.1145/2464576.2464655
Li X, Hirasawa K. Extended rule-based genetic network programming. In GECCO 2013 - Proceedings of the 2013 Genetic and Evolutionary Computation Conference Companion. 2013. p. 155-156 https://doi.org/10.1145/2464576.2464655
Li, Xianneng ; Hirasawa, Kotaro. / Extended rule-based genetic network programming. GECCO 2013 - Proceedings of the 2013 Genetic and Evolutionary Computation Conference Companion. 2013. pp. 155-156
@inproceedings{43b1ad361f174b05be78e2330608382d,
title = "Extended rule-based genetic network programming",
abstract = "Recent advances in rule-based systems, i.e., Learning Classifier Systems (LCSs), have shown their sequential decisionmaking ability with a generalization property. In this paper, a novel LCS named eXtended rule-based Genetic Network Programming (XrGNP) is proposed. Different from most of the current LCSs, the rules are represented and discovered through a graph-based evolutionary algorithm GNP, which consequently has the distinct expression ability to model and evolve the {"}if-then{"} decision-making rules. Experiments on a benchmark multi-step problem (so-called Reinforcement Learning problem) demonstrate its effectiveness.",
keywords = "Genetic network programming, Learning classifier systems, XrGNP",
author = "Xianneng Li and Kotaro Hirasawa",
year = "2013",
doi = "10.1145/2464576.2464655",
language = "English",
isbn = "9781450319645",
pages = "155--156",
booktitle = "GECCO 2013 - Proceedings of the 2013 Genetic and Evolutionary Computation Conference Companion",

}

TY - GEN

T1 - Extended rule-based genetic network programming

AU - Li, Xianneng

AU - Hirasawa, Kotaro

PY - 2013

Y1 - 2013

N2 - Recent advances in rule-based systems, i.e., Learning Classifier Systems (LCSs), have shown their sequential decisionmaking ability with a generalization property. In this paper, a novel LCS named eXtended rule-based Genetic Network Programming (XrGNP) is proposed. Different from most of the current LCSs, the rules are represented and discovered through a graph-based evolutionary algorithm GNP, which consequently has the distinct expression ability to model and evolve the "if-then" decision-making rules. Experiments on a benchmark multi-step problem (so-called Reinforcement Learning problem) demonstrate its effectiveness.

AB - Recent advances in rule-based systems, i.e., Learning Classifier Systems (LCSs), have shown their sequential decisionmaking ability with a generalization property. In this paper, a novel LCS named eXtended rule-based Genetic Network Programming (XrGNP) is proposed. Different from most of the current LCSs, the rules are represented and discovered through a graph-based evolutionary algorithm GNP, which consequently has the distinct expression ability to model and evolve the "if-then" decision-making rules. Experiments on a benchmark multi-step problem (so-called Reinforcement Learning problem) demonstrate its effectiveness.

KW - Genetic network programming

KW - Learning classifier systems

KW - XrGNP

UR - http://www.scopus.com/inward/record.url?scp=84882420291&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84882420291&partnerID=8YFLogxK

U2 - 10.1145/2464576.2464655

DO - 10.1145/2464576.2464655

M3 - Conference contribution

AN - SCOPUS:84882420291

SN - 9781450319645

SP - 155

EP - 156

BT - GECCO 2013 - Proceedings of the 2013 Genetic and Evolutionary Computation Conference Companion

ER -