Genetic network programming with actor-critic and its application

Shingo Mabu, Kotaro Hirasawa, Takayuki Furuzuki

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

Abstract

A new graph-based evolutionary algorithm named "Genetic Network Programming, GNP" was proposed. GNP represents its solutions as graph structures, which can improve the expression ability and performance. And then, GNP with Reinforcement Learning (GNP with RL) has been proposed in order to search for solutions efficiently. GNP with RL can use the current information (state and reward) and change its programs during task execution, so it has an advantage over the evolution based algorithms in case much information can be obtained during task execution. In this paper, GNP with Actor-Critic (GNP-AC) which is a new type of GNP with RL is proposed. Originally, GNP deals with discrete information (ex. right, left, etc.), but GNP with AC aims to deal with continuous information (ex. the sensor value is "32"). The proposed method is applied to the controller of the Khepera simulator and its performance is evaluated.

Original languageEnglish
Title of host publicationProceedings of the SICE Annual Conference
Pages3635-3640
Number of pages6
Publication statusPublished - 2005
EventSICE Annual Conference 2005 - Okayama
Duration: 2005 Aug 82005 Aug 10

Other

OtherSICE Annual Conference 2005
CityOkayama
Period05/8/805/8/10

Fingerprint

Reinforcement learning
Evolutionary algorithms
Simulators
Controllers
Sensors

Keywords

  • Actor-Critic
  • Evolutionary Computation
  • Genetic Network Programming
  • Khepera robot
  • Reinforcement Learning

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Mabu, S., Hirasawa, K., & Furuzuki, T. (2005). Genetic network programming with actor-critic and its application. In Proceedings of the SICE Annual Conference (pp. 3635-3640)

Genetic network programming with actor-critic and its application. / Mabu, Shingo; Hirasawa, Kotaro; Furuzuki, Takayuki.

Proceedings of the SICE Annual Conference. 2005. p. 3635-3640.

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

Mabu, S, Hirasawa, K & Furuzuki, T 2005, Genetic network programming with actor-critic and its application. in Proceedings of the SICE Annual Conference. pp. 3635-3640, SICE Annual Conference 2005, Okayama, 05/8/8.
Mabu S, Hirasawa K, Furuzuki T. Genetic network programming with actor-critic and its application. In Proceedings of the SICE Annual Conference. 2005. p. 3635-3640
Mabu, Shingo ; Hirasawa, Kotaro ; Furuzuki, Takayuki. / Genetic network programming with actor-critic and its application. Proceedings of the SICE Annual Conference. 2005. pp. 3635-3640
@inproceedings{e2fb21e6920b41f09c239437dac7a909,
title = "Genetic network programming with actor-critic and its application",
abstract = "A new graph-based evolutionary algorithm named {"}Genetic Network Programming, GNP{"} was proposed. GNP represents its solutions as graph structures, which can improve the expression ability and performance. And then, GNP with Reinforcement Learning (GNP with RL) has been proposed in order to search for solutions efficiently. GNP with RL can use the current information (state and reward) and change its programs during task execution, so it has an advantage over the evolution based algorithms in case much information can be obtained during task execution. In this paper, GNP with Actor-Critic (GNP-AC) which is a new type of GNP with RL is proposed. Originally, GNP deals with discrete information (ex. right, left, etc.), but GNP with AC aims to deal with continuous information (ex. the sensor value is {"}32{"}). The proposed method is applied to the controller of the Khepera simulator and its performance is evaluated.",
keywords = "Actor-Critic, Evolutionary Computation, Genetic Network Programming, Khepera robot, Reinforcement Learning",
author = "Shingo Mabu and Kotaro Hirasawa and Takayuki Furuzuki",
year = "2005",
language = "English",
pages = "3635--3640",
booktitle = "Proceedings of the SICE Annual Conference",

}

TY - GEN

T1 - Genetic network programming with actor-critic and its application

AU - Mabu, Shingo

AU - Hirasawa, Kotaro

AU - Furuzuki, Takayuki

PY - 2005

Y1 - 2005

N2 - A new graph-based evolutionary algorithm named "Genetic Network Programming, GNP" was proposed. GNP represents its solutions as graph structures, which can improve the expression ability and performance. And then, GNP with Reinforcement Learning (GNP with RL) has been proposed in order to search for solutions efficiently. GNP with RL can use the current information (state and reward) and change its programs during task execution, so it has an advantage over the evolution based algorithms in case much information can be obtained during task execution. In this paper, GNP with Actor-Critic (GNP-AC) which is a new type of GNP with RL is proposed. Originally, GNP deals with discrete information (ex. right, left, etc.), but GNP with AC aims to deal with continuous information (ex. the sensor value is "32"). The proposed method is applied to the controller of the Khepera simulator and its performance is evaluated.

AB - A new graph-based evolutionary algorithm named "Genetic Network Programming, GNP" was proposed. GNP represents its solutions as graph structures, which can improve the expression ability and performance. And then, GNP with Reinforcement Learning (GNP with RL) has been proposed in order to search for solutions efficiently. GNP with RL can use the current information (state and reward) and change its programs during task execution, so it has an advantage over the evolution based algorithms in case much information can be obtained during task execution. In this paper, GNP with Actor-Critic (GNP-AC) which is a new type of GNP with RL is proposed. Originally, GNP deals with discrete information (ex. right, left, etc.), but GNP with AC aims to deal with continuous information (ex. the sensor value is "32"). The proposed method is applied to the controller of the Khepera simulator and its performance is evaluated.

KW - Actor-Critic

KW - Evolutionary Computation

KW - Genetic Network Programming

KW - Khepera robot

KW - Reinforcement Learning

UR - http://www.scopus.com/inward/record.url?scp=33645283613&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33645283613&partnerID=8YFLogxK

M3 - Conference contribution

SP - 3635

EP - 3640

BT - Proceedings of the SICE Annual Conference

ER -