You are here

An Adaptive Multiobjective Approach to Evolving ART Architectures

TitleAn Adaptive Multiobjective Approach to Evolving ART Architectures
Publication TypeJournal Article
Year of Publication2010
AuthorsKaylani A, Georgiopoulos M, Mollaghasemi M, Anagnostopoulos GC, Sentelle C, Zhong M
JournalNeural Networks, IEEE Transactions on
Volume21
Pagination529-550
Date PublishedApril
ISSN1045-9227
Keywordsadaptive multiobjective approach, adaptive resonance theory, Algorithms, ART architecture, ART neural nets, ART neural network architecture, Artificial Intelligence, ARTMAP, Automated, category proliferation, classification, classification problem, Computer Simulation, ellipsoidal ARTMAP, evolutionary computation, fuzzy ARTMAP, Fuzzy Logic, fuzzy neural nets, Gaussian ARTMAP, genetic algorithms (GAs), genetic operators, Humans, Machine learning, multiobjective evolutionary approach, multiobjective optimization, neural net architecture, Neural Networks (Computer), Normal Distribution, optimisation, pattern recognition
Abstract

In this paper, we present the evolution of adaptive resonance theory (ART) neural network architectures (classifiers) using a multiobjective optimization approach. In particular, we propose the use of a multiobjective evolutionary approach to simultaneously evolve the weights and the topology of three well-known ART architectures; fuzzy ARTMAP (FAM), ellipsoidal ARTMAP (EAM), and Gaussian ARTMAP (GAM). We refer to the resulting architectures as MO-GFAM, MO-GEAM, and MO-GGAM, and collectively as MO-GART. The major advantage of MO-GART is that it produces a number of solutions for the classification problem at hand that have different levels of merit [accuracy on unseen data (generalization) and size (number of categories created)]. MO-GART is shown to be more elegant (does not require user intervention to define the network parameters), more effective (of better accuracy and smaller size), and more efficient (faster to produce the solution networks) than other ART neural network architectures that have appeared in the literature. Furthermore, MO-GART is shown to be competitive with other popular classifiers, such as classification and regression tree (CART) and support vector machines (SVMs).

DOI10.1109/TNN.2009.2037813

Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer