Comparative analysis of fuzzy ART and ART-2A network clustering performance

Frank, T.; Kraiss, Karl-Friedrich; Kuhlen, Torsten

doi:10.1109/72.668896

Cited by 105 publications

(55 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The input patterns are to be well aligned and normalized while presenting to such type of MLP leading to no built in variance with respect to local distortions and translations [20]. Various neural network classifiers are explained in [18,[21][22][23][24][25][26]. A convolutional neural network(CNN) tries to solve the problems of MLP by extracting local features and combining them subsequently to perform the detection or recognition.…”

Section: Introductionmentioning

confidence: 99%

A novel training algorithm for convolutional neural network

Anuse¹,

Vyas²

2016

Complex Intell. Syst.

View full text Add to dashboard Cite

Many machine learning softwares are available which help the researchers to accomplish various tasks. These software packages have various conventional algorithms which perform well if the training and test data are independent and identically distributed. However, this might not be the case in the real world. The training data may not be available at one time. In the case of neural networks, the architecture has to be retrained with new data that are made available subsequently. In this paper, we present a novel training algorithm which can avoid complete retraining of any neural network architecture meant for visual pattern recognition. To show the utility of the algorithm, we have investigated the performance of convolutional neural network (CNN) architecture for a face recognition task under transfer learning. The proposed training algorithm may be used for enhancing the utility of machine learning software by providing researchers with an approach that can reduce the training time under transfer learning.

show abstract

Section: Introductionmentioning

confidence: 99%

A novel training algorithm for convolutional neural network

Anuse¹,

Vyas²

2016

Complex Intell. Syst.

View full text Add to dashboard Cite

show abstract

“…Além disso, este modelo de RNA foi desenvolvido baseando-se no modelo ART1, mas adaptado ao tratamento de valores reais pertencentes ao intervalo [0, 1] (i.e, valores fuzzy) (Carpenter et al, 1991b). Mais detalhes a respeito deste modelo também podem ser encontrados nos trabalhos de (Frank et al, 1998) e (Baraldi and Bionda, 1999).…”

Section: Fuzzy Artunclassified

“…O algoritmo segue avaliando se o neurõnio vencedor, em F2, é suficientemente similar ao padrão de entrada x; se for verdadeiro, a rede entra em processo de atualização de pesos (ressonância). (Carpenter et al, 1991b) (Frank et al, 1998).…”

Section: Descrição Do Treinamento Da Rede Fuzzy Artunclassified

“…O modelo de RNA ART2-A também foi analisado nos seguintes trabalhos: (Peper et al, 1993) e (Frank et al, 1998). No primeiro trabalho, o autores observam que 10 ciclos de treinamento são suficientes para estabilizar a rede em seus experimentos.…”

Section: Considerações Sobre Os Modelos Art Estudadosunclassified

“…Dentre várias análises existentes na literatura a respeito do modelo Fuzzy ART, foram considerados, neste presente trabalho, os seguintes: (Frank et al, 1998) e (Baraldi and Alpaydin, 2002a). As conclusões a respeito das potenciais fraquezas do modelo podem ser resumidas nos seguintes itens:…”

Section: Considerações Sobre Os Modelos Art Estudadosunclassified

See 2 more Smart Citations