Genetic network programming with automatically generated variable size macro nodes

Hiroshi Nakagoe, Kotaro Hirasawa, Takayuki Furuzuki

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

4 Citations (Scopus)

Abstract

Genetic Network Programming (GNP) has directed graph structures as genes, which is extended from other evolutionary computations such as Genetic Algorithm (GA) and Genetic Programming (GP). Generally, macroinstructions are introduced as sub-routines, function localization and so on. Previously, we have introduced the structure of macroinstructions in GNP named Automatically Generated Macro Nodes (AGMs) for reducing the time of evolution efficiently, and showed that macroinstructions are useful to acquire good performances. But the AGMs have fixed number of nodes, and it is found that the effectiveness of evolution of macroinstructions depends on the main program calling them and initialized parameters. Accordingly in this paper, new AGMs are introduced to improve their performances further more by the mechanism of varying the size of AGMs, which are named variable size AGMs. This is the mechanism to add and delete nodes according to necessity. In the simulations, comparisons between GNP program only, GNP with conventional AGMs and GNP with variable size AGMs are carried out using the tile world. Simulation results show that the proposed method is better compared with conventional GNP and GNP with conventional AGMs, And also it is clarified that the node transition rules obtained by new AGMs show the generalized rules able to deal with unknown environments.

Original languageEnglish
Title of host publicationProceedings of the 2004 Congress on Evolutionary Computation, CEC2004
Pages713-719
Number of pages7
Volume1
Publication statusPublished - 2004
EventProceedings of the 2004 Congress on Evolutionary Computation, CEC2004 - Portland, OR
Duration: 2004 Jun 192004 Jun 23

Other

OtherProceedings of the 2004 Congress on Evolutionary Computation, CEC2004
CityPortland, OR
Period04/6/1904/6/23

Fingerprint

Computer programming
Macros
Genetic programming
Directed graphs
Tile
Evolutionary algorithms
Genes
Genetic algorithms

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Nakagoe, H., Hirasawa, K., & Furuzuki, T. (2004). Genetic network programming with automatically generated variable size macro nodes. In Proceedings of the 2004 Congress on Evolutionary Computation, CEC2004 (Vol. 1, pp. 713-719)

Genetic network programming with automatically generated variable size macro nodes. / Nakagoe, Hiroshi; Hirasawa, Kotaro; Furuzuki, Takayuki.

Proceedings of the 2004 Congress on Evolutionary Computation, CEC2004. Vol. 1 2004. p. 713-719.

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

Nakagoe, H, Hirasawa, K & Furuzuki, T 2004, Genetic network programming with automatically generated variable size macro nodes. in Proceedings of the 2004 Congress on Evolutionary Computation, CEC2004. vol. 1, pp. 713-719, Proceedings of the 2004 Congress on Evolutionary Computation, CEC2004, Portland, OR, 04/6/19.
Nakagoe H, Hirasawa K, Furuzuki T. Genetic network programming with automatically generated variable size macro nodes. In Proceedings of the 2004 Congress on Evolutionary Computation, CEC2004. Vol. 1. 2004. p. 713-719
Nakagoe, Hiroshi ; Hirasawa, Kotaro ; Furuzuki, Takayuki. / Genetic network programming with automatically generated variable size macro nodes. Proceedings of the 2004 Congress on Evolutionary Computation, CEC2004. Vol. 1 2004. pp. 713-719
@inproceedings{1ba12b94b4c04efbbe948ede4f1d2ee4,
title = "Genetic network programming with automatically generated variable size macro nodes",
abstract = "Genetic Network Programming (GNP) has directed graph structures as genes, which is extended from other evolutionary computations such as Genetic Algorithm (GA) and Genetic Programming (GP). Generally, macroinstructions are introduced as sub-routines, function localization and so on. Previously, we have introduced the structure of macroinstructions in GNP named Automatically Generated Macro Nodes (AGMs) for reducing the time of evolution efficiently, and showed that macroinstructions are useful to acquire good performances. But the AGMs have fixed number of nodes, and it is found that the effectiveness of evolution of macroinstructions depends on the main program calling them and initialized parameters. Accordingly in this paper, new AGMs are introduced to improve their performances further more by the mechanism of varying the size of AGMs, which are named variable size AGMs. This is the mechanism to add and delete nodes according to necessity. In the simulations, comparisons between GNP program only, GNP with conventional AGMs and GNP with variable size AGMs are carried out using the tile world. Simulation results show that the proposed method is better compared with conventional GNP and GNP with conventional AGMs, And also it is clarified that the node transition rules obtained by new AGMs show the generalized rules able to deal with unknown environments.",
author = "Hiroshi Nakagoe and Kotaro Hirasawa and Takayuki Furuzuki",
year = "2004",
language = "English",
isbn = "0780385152",
volume = "1",
pages = "713--719",
booktitle = "Proceedings of the 2004 Congress on Evolutionary Computation, CEC2004",

}

TY - GEN

T1 - Genetic network programming with automatically generated variable size macro nodes

AU - Nakagoe, Hiroshi

AU - Hirasawa, Kotaro

AU - Furuzuki, Takayuki

PY - 2004

Y1 - 2004

N2 - Genetic Network Programming (GNP) has directed graph structures as genes, which is extended from other evolutionary computations such as Genetic Algorithm (GA) and Genetic Programming (GP). Generally, macroinstructions are introduced as sub-routines, function localization and so on. Previously, we have introduced the structure of macroinstructions in GNP named Automatically Generated Macro Nodes (AGMs) for reducing the time of evolution efficiently, and showed that macroinstructions are useful to acquire good performances. But the AGMs have fixed number of nodes, and it is found that the effectiveness of evolution of macroinstructions depends on the main program calling them and initialized parameters. Accordingly in this paper, new AGMs are introduced to improve their performances further more by the mechanism of varying the size of AGMs, which are named variable size AGMs. This is the mechanism to add and delete nodes according to necessity. In the simulations, comparisons between GNP program only, GNP with conventional AGMs and GNP with variable size AGMs are carried out using the tile world. Simulation results show that the proposed method is better compared with conventional GNP and GNP with conventional AGMs, And also it is clarified that the node transition rules obtained by new AGMs show the generalized rules able to deal with unknown environments.

AB - Genetic Network Programming (GNP) has directed graph structures as genes, which is extended from other evolutionary computations such as Genetic Algorithm (GA) and Genetic Programming (GP). Generally, macroinstructions are introduced as sub-routines, function localization and so on. Previously, we have introduced the structure of macroinstructions in GNP named Automatically Generated Macro Nodes (AGMs) for reducing the time of evolution efficiently, and showed that macroinstructions are useful to acquire good performances. But the AGMs have fixed number of nodes, and it is found that the effectiveness of evolution of macroinstructions depends on the main program calling them and initialized parameters. Accordingly in this paper, new AGMs are introduced to improve their performances further more by the mechanism of varying the size of AGMs, which are named variable size AGMs. This is the mechanism to add and delete nodes according to necessity. In the simulations, comparisons between GNP program only, GNP with conventional AGMs and GNP with variable size AGMs are carried out using the tile world. Simulation results show that the proposed method is better compared with conventional GNP and GNP with conventional AGMs, And also it is clarified that the node transition rules obtained by new AGMs show the generalized rules able to deal with unknown environments.

UR - http://www.scopus.com/inward/record.url?scp=4344658187&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=4344658187&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:4344658187

SN - 0780385152

SN - 9780780385153

VL - 1

SP - 713

EP - 719

BT - Proceedings of the 2004 Congress on Evolutionary Computation, CEC2004

ER -