Theta-Fuzzy Associative Memories (Theta-FAMs)

Autor(es)

Esmi, Estevao ^[1] ; Sussner, Peter ^[1] ; Bustince, Humberto ^[2] ; Fernandez, Javier ^[2]

Afiliação do(s) autor(es)

^[1] Univ Estadual Campinas, Dept Appl Math, BR-13081970 Campinas, SP - Brazil

^[2] Univ Publ Navarra, Pamplona 31006 - Spain

• Tipo de documento: Artigo Científico
• Fonte: IEEE TRANSACTIONS ON FUZZY SYSTEMS; v. 23, n. 2, p. 313-326, APR 2015.
• Citações Web of Science: 11

Resumo

Most fuzzy associative memories (FAMs) in the literature correspond to neural networks with a single layer of weights that distributively contains the information on associations to be stored. The main applications of these types of associative memory can be found in fuzzy rule-based systems. In contrast, T-fuzzy associative memories (T-FAMs) represent parametrized fuzzy neural networks with a hidden layer and these FAM models extend (dual) S-FAMs and SM-FAMs based on fuzzy subsethood and similarity measures. In this paper, we provide theoretical results concerning the storage capacity and error correction capability of T-FAMs. In addition, we introduce a training algorithm for T-FAMs and we compare the error rates produced by T-FAMs and some well-known classifiers in some benchmark classification problems that are available on the internet. Finally, we apply T-FAMs to a problem of vision-based self-localization in mobile robotics. (AU)

• Processo FAPESP: 09/16284-2 - Estimação de Mapeamentos entre Reticulados Utilizando Neurocomputação (Fuzzy) para Reconhecimento de Padrões
• Beneficiário: Estevão Esmi Laureano
• Modalidade de apoio: Bolsas no Brasil - Doutorado
• Processo FAPESP: 11/10014-3 - Métodos de inteligência computacional e processamento de imagens baseados em morfologia matemática e álgebra de reticulados
• Beneficiário: Peter Sussner
• Modalidade de apoio: Auxílio à Pesquisa - Regular