A Cognitive Architecture Based on a Learning Classifier System with Spiking Classifiers

Howard, David; Howard, Gerard David; Bull, Larry; Lanzi, Pier Luca

Abstract

© 2015, Springer Science+Business Media New York. Learning classifier systems (LCS) are population-based reinforcement learners that were originally designed to model various cognitive phenomena. This paper presents an explicitly cognitive LCS by using spiking neural networks as classifiers, providing each classifier with a measure of temporal dynamism. We employ a constructivist model of growth of both neurons and synaptic connections, which permits a genetic algorithm to automatically evolve sufficiently-complex neural structures. The spiking classifiers are coupled with a temporally-sensitive reinforcement learning algorithm, which allows the system to perform temporal state decomposition by appropriately rewarding “macro-actions”, created by chaining together multiple atomic actions. The combination of temporal reinforcement learning and neural information processing is shown to outperform benchmark neural classifier systems, and successfully solve a robotic navigation task.

Citation

Howard, D., Howard, G. D., Bull, L., & Lanzi, P. L. (2016). A Cognitive Architecture Based on a Learning Classifier System with Spiking Classifiers. Neural Processing Letters, 44(1), 125-147. https://doi.org/10.1007/s11063-015-9451-4

• Journal Article Type: Article
• Online Publication Date: Jun 26, 2015
• Publication Date: Aug 1, 2016
• Journal: Neural Processing Letters
• Print ISSN: 1370-4621
• Electronic ISSN: 1573-773X
• Publisher: Springer Verlag
• Peer Reviewed: Peer Reviewed
• Volume: 44
• Issue: 1
• Pages: 125-147
• DOI: https://doi.org/10.1007/s11063-015-9451-4
• Keywords: evolution, neural network
• Public URL: https://uwe-repository.worktribe.com/output/910717
• Publisher URL: http://dx.doi.org/10.1007/s11063-015-9451-4
• Additional Information: Additional Information : The final publication is available at Springer via http://dx.doi.org/10.1007/s11063-015-9451-4