Use of a connection-selection scheme in neural XCSF

Howard, G. D., Bull, L. and Lanzi, P.-L. (2010) Use of a connection-selection scheme in neural XCSF. In: (2010) Learning Classifier Systems. Springer Verlag, pp. 87-106. ISBN 9783642175077 Available from: http://eprints.uwe.ac.uk/20722

Full text not available from this repository

Publisher's URL: http://www.springerlink.com

Abstract/Description

XCSF is a modern form of Learning Classifier System (LCS) that has proven successful in a number of problem domains. In this paper we exploit the modular nature of XCSF to include a number of extensions, namely a neural classifier representation, self-adaptive mutation rates and neural constructivism. It is shown that, via constructivism, appropriate internal rule complexity emerges during learning. It is also shown that self-adaptation allows this rule complexity to emerge at a rate controlled by the learner. We evaluate this system on both discrete and continuous-valued maze environments. The main contribution of this work is the implementation of a feature selection derivative (termed connection selection), which is applied to modify network connectivity patterns. We evaluate the effect of connection selection, in terms of both solution size and system performance, on both discrete and continuous-valued

Item Type:Book Section
Uncontrolled Keywords:computation by abstract devices, computer science, artificial intelligence, robotics, algorithm analysis, problem complexity, database management, information systems applications, internet
Faculty/Department:Faculty of Environment and Technology > Department of Computer Science and Creative Technologies
ID Code:20722
Deposited By: M. Clarke
Deposited On:15 Jul 2013 14:44
Last Modified:15 Nov 2016 22:20

Request a change to this item