Consideration of multiple objectives in neural learning classifier systems

Bull, Larry; Studley, Matthew

Abstract

© Springer-Verlag Berlin Heidelberg 2002. For effective use in a number of problem domains Learning Classifier Systems must be able to manage multiple objectives. This paper explicitly considers the case of developing the controller for a simulated mobile autonomous robot which must achieve a given task whilst maintaining sufficient battery power. A form of Learning Classifier System in which each rule is represented by an artificial neural network is used. Results are presented which show it is possible to solve both objectives when the energy level is presented as an input along with sensor data. A more realistic, and hence more complex, version of the basic scenario is then investigated.

Citation

Bull, L., & Studley, M. (2002). Consideration of multiple objectives in neural learning classifier systems. In J. J. Merelo, P. Adamidis, & H. Beyer (Eds.), Parallel Problem Solving from Nature—PPSN VII. , (549-557). https://doi.org/10.1007/3-540-45712-7_53

• Conference Name: Parallel Problem Solving from Nature—PPSN VII
• Conference Location: Granada, Spain
• Start Date: Sep 7, 2002
• End Date: Sep 11, 2002
• Publication Date: Jan 1, 2002
• Publisher: Springer Verlag
• Peer Reviewed: Not Peer Reviewed
• Pages: 549-557
• Series Title: Lecture Notes in Computer Science
• Series Number: 2439
• Series ISSN: 0302-9743
• Book Title: Parallel Problem Solving from Nature—PPSN VII
• ISBN: 9783540441397
• DOI: https://doi.org/10.1007/3-540-45712-7_53
• Keywords: multiple objectives, neural learning classifier systems
• Public URL: https://uwe-repository.worktribe.com/output/1082239
• Publisher URL: http://dx.doi.org/10.1007/3-540-45712-7_53