Modeling UCS as a mixture of experts

Authors:
Narayanan Unny Edakunni;Tim Kovacs;Gavin Brown;James A.R. Marshall
Affiliations:
University of Bristol, Bristol, United Kingdom;University of Bristol, Bristol, United Kingdom;University of Manchester, Manchester, United Kingdom;University of Bristol, Bristol, United Kingdom
Venue:
Proceedings of the 11th Annual conference on Genetic and evolutionary computation
Year:
2009

Citing 5
Cited 3

Accuracy-based learning classifier systems: models, analysis and applications to classification tasks

Evolutionary Computation
Managing Diversity in Regression Ensembles

The Journal of Machine Learning Research
Bayesian estimation of rule accuracy in UCS

Proceedings of the 9th annual conference companion on Genetic and evolutionary computation
Adaptive mixtures of local experts

Neural Computation
Design and Analysis of Learning Classifier Systems: A Probabilistic Approach (Studies in Computational Intelligence)

Design and Analysis of Learning Classifier Systems: A Probabilistic Approach (Studies in Computational Intelligence)

Accuracy exponentiation in UCS and its effect on voting margins

Proceedings of the 13th annual conference on Genetic and evolutionary computation
Online, GA based mixture of experts: a probabilistic model of ucs

Proceedings of the 13th annual conference on Genetic and evolutionary computation
Modelling the initialisation stage of the ALKR representation for discrete domains and GABIL encoding

Proceedings of the 13th annual conference on Genetic and evolutionary computation

Quantified Score

Hi-index	0.00

Visualization

Abstract

We present a probabilistic formulation of UCS (a sUpervised Classifier System). UCS is shown to be a special case of mixture of experts where the experts are learned independently and later combined during prediction. In this work, we develop the links between the constituent components of UCS and a mixture of experts, thus lending UCS a strong analytical background. We find during our analysis that mixture of experts is a more generic formulation of UCS and possesses more generalization capability and flexibility than UCS, which is also verified using empirical evaluations. This is the first time that a simple probabilistic model has been proposed for UCS and we believe that this work will form a useful tool to analyse Learning Classifier Systems and gain useful insights into their working.