Artificial life simulation using marker-based encoding

Dara Curran, Colm O'Riordan

Research output: Chapter in Book or Conference Publication/ProceedingConference Publicationpeer-review

5 Citations (Scopus)

Abstract

This paper describes the design of an artificial life simulator. The simulator uses a genetic algorithm to evolve a population of neural networks to solve a presented set of problems. The simulator has been designed to facilitate experimentation in combining different forms of learning (evolutionary algorithms and neural networks). We present results obtained in simulations where the population is evolved to solve certain problems. The simulations are designed to show the population's progress when presented with problems of increasing difficulty using evolutionary algorithms and neural networks both individually and in combination. In mapping the structure of a neural network to an encoding suitable for the genetic algorithm, marker-based encoding is used.

Original languageEnglish
Title of host publicationProceedings of the International Conference on Artificial Intelligence IC-AI 2003
EditorsH.R. Arabnia, R. Joshua, Y. Mun, H.R. Arabnia, R. Joshua, Y. Mun
Pages665-668
Number of pages4
Publication statusPublished - 2003
EventProceedings of the International Conference on Artificial Intelligence, IC-AI 2003 - Las Vegas, NV, United States
Duration: 23 Jun 200326 Jun 2003

Publication series

NameProceedings of the International Conference on Artificial Intelligence IC-AI 2003
Volume2

Conference

ConferenceProceedings of the International Conference on Artificial Intelligence, IC-AI 2003
Country/TerritoryUnited States
CityLas Vegas, NV
Period23/06/0326/06/03

Keywords

  • Artificial life
  • Genetic algorithms
  • Neural networks
  • Simulation

Fingerprint

Dive into the research topics of 'Artificial life simulation using marker-based encoding'. Together they form a unique fingerprint.

Cite this