Brave Mobots Use Representation: Emergence ofRepresentation in Fight-or-Flight Learning

  • Authors:
  • Chris Thornton

  • Affiliations:
  • Cognitive and Computing Sciences, University of Sussex, Brighton, BN1 9QH, U.K.

  • Venue:
  • Minds and Machines
  • Year:
  • 1997

Quantified Score

Hi-index 0.00

Visualization

Abstract

The paper uses ideas from Machine Learning, ArtificialIntelligence and Genetic Algorithms to provide a model ofthe development of a ’fight-or-flight‘ response in asimulated agent. The modelled developmentprocess involves (simulated) processes of evolution,learning and representation development. The main value ofthe model is that it provides an illustration of how simplelearning processes may lead to the formation of structureswhich can be given a representational interpretation. Italso shows how these may form the infrastructure forclosely-coupled agent/environment interaction.