Title :
Analysis of the backpropagation algorithm using linear algebra
Author :
De Sousa, Celso André Rodrigues
Author_Institution :
Inst. of Math. & Comput. Sci. (ICMC), Univ. of Sao Paulo (USP), Sao Carlos, Brazil
Abstract :
Multilayer perceptrons (MLPs) are feed-forward artificial neural networks with high theoretical basis. The most popular algorithm to train MLPs is the backpropagation algorithm, which can be seen as a consistent nonparametric least squares regression estimator. This algorithm is reformulated in this paper using linear algebra, providing theoretical basis for further studies.
Keywords :
backpropagation; feedforward neural nets; learning (artificial intelligence); least squares approximations; linear algebra; multilayer perceptrons; regression analysis; MLP training; backpropagation algorithm; consistent nonparametric least squares regression estimator; feed-forward artificial neural networks; linear algebra; multilayer perceptrons; Backpropagation algorithms; Biological neural networks; Logistics; Neurons; Training; Vectors;
Conference_Titel :
Neural Networks (IJCNN), The 2012 International Joint Conference on
Conference_Location :
Brisbane, QLD
Print_ISBN :
978-1-4673-1488-6
Electronic_ISBN :
2161-4393
DOI :
10.1109/IJCNN.2012.6252364
