Genetic Network Programming with reinforcement learning for generating agent behavior in the benchmark problems

Shingo Mabu, Kotaro Hirasawa, Jinglu Hu

Research output: Contribution to conferencePaper

Abstract

A new graph-based evolutionary algorithm named "Genetic Network Programming, GNP" has been proposed. GNP represents its solutions as graph structures which have distinguished expression ability. In this paper, we propose GNP with Reinforcement Learning. Evolutionary algorithm of GNP makes a very compact graph structure and Reinforcement Learning of GNP improves search speed for solutions.

Original languageEnglish
Pages605-610
Number of pages6
Publication statusPublished - 2004 Dec 1
EventSICE Annual Conference 2004 - Sapporo, Japan
Duration: 2004 Aug 42004 Aug 6

Conference

ConferenceSICE Annual Conference 2004
CountryJapan
CitySapporo
Period04/8/404/8/6

Keywords

  • Agent
  • Genetic Programming
  • Graph structure
  • Reinforcement Learning

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Computer Science Applications
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Genetic Network Programming with reinforcement learning for generating agent behavior in the benchmark problems'. Together they form a unique fingerprint.

  • Cite this

    Mabu, S., Hirasawa, K., & Hu, J. (2004). Genetic Network Programming with reinforcement learning for generating agent behavior in the benchmark problems. 605-610. Paper presented at SICE Annual Conference 2004, Sapporo, Japan.