Redes de Função de Base Radial Utilizando a Função Lambert-Tsallis

Detalhes bibliográficos
Ano de defesa: 2020
Autor(a) principal: Silva, Jorge Lucas Mouta da
Orientador(a): Não Informado pela instituição
Banca de defesa: Não Informado pela instituição
Tipo de documento: Dissertação
Tipo de acesso: Acesso aberto
Idioma: por
Instituição de defesa: Não Informado pela instituição
Programa de Pós-Graduação: Não Informado pela instituição
Departamento: Não Informado pela instituição
País: Não Informado pela instituição
Palavras-chave em Português:
Link de acesso: http://www.repositorio.ufc.br/handle/riufc/56529
Resumo: Artificial neural networks are computational models based on the central nervous system that can approximate unknown functions, so they perform pattern recognition very well. There are different types of neural networks. This dissertation works exclusively with Radial Base Function Networks (RBFN). The main element of an RBFN is exactly the radial base function used. The present work proposes a new radial base function based on the recently created Lambert-Tsallis function. The RBFN using the Lambert-Tsallis function was then used in two tasks: 1) As a classifier capable of discriminating between entangled and disentangled states of two qubits. 2) As a probability density function estimator based on a random sequence of numbers. In the first case, quantum states with entanglement greater than 0.1 were correctly classified by the proposed RBFN in at least 97% of cases. In the second case, the RBFN estimated the probability density functions of two sampled data sets with good precision, according to the Normal and Cauchy distributions. In both cases the results obtained were compared with the results obtained by another RBFN using the q-Gaussian function with different values of q as the radial basis function.