| Texto completo | |
| Autor(es): |
Sussner, Peter
;
Esmi, Estevao
;
Florindo, Joao
Número total de Autores: 3
|
| Tipo de documento: | Artigo Científico |
| Fonte: | 2019 IEEE INTERNATIONAL CONFERENCE ON FUZZY SYSTEMS, FUZZ-IEEE; v. N/A, p. 6-pg., 2019-01-01. |
| Resumo | |
A (weighted) Theta-fuzzy associative memory can be viewed as single hidden layer neural network whose inputs are drawn from an arbitrary bounded lattice L. The xi th hidden node applies a function Theta(xi) : L -> [0; 1] to an input A is an element of L. In this paper, we present a new algorithm for designing Theta-FAMs. Roughly speaking, this algorithm consists of the following two stages: 1) Construction of a set of functions Theta(xi) : L -> [0; 1] where xi = 1, ...,, p; 2) Optimization of the weights. Our new algorithm differs from the previous algorithm for tunable equivalence fuzzy associative memories. Instead of merely extracting a subset of functions Theta(xi) : L -> [0; 1] from a given set, we generate a new set of functions on the basis of the partial ordering of L. The paper includes some experimental results in a set of benchmark classification problems. We also apply a combination of the resulting Theta-FAM and a deep convolutional neural network to a problem of image texture classification and compare the classification performance of our approach with the ones of some state-of-the-art methods from the literature. (AU) | |
| Processo FAPESP: | 18/13657-1 - Algumas abordagens de computação em reticulados a inteligência computacional, processamento e análise de imagens |
| Beneficiário: | Peter Sussner |
| Modalidade de apoio: | Auxílio à Pesquisa - Regular |
| Processo FAPESP: | 16/16060-0 - Reconhecimento de Padrões em Imagens Baseado em Sistemas Complexos |
| Beneficiário: | Joao Batista Florindo |
| Modalidade de apoio: | Auxílio à Pesquisa - Regular |
| Processo FAPESP: | 16/26040-7 - Cálculo diferencial e integral baseado aritmética de números fuzzy interativos |
| Beneficiário: | Estevão Esmi Laureano |
| Modalidade de apoio: | Auxílio à Pesquisa - Regular |