Title :
A simple neuron feature detection
Author :
Hambaba, Mohamed L.
Author_Institution :
Dept. of Electr. Eng. & Comput. Sci., Stevens Inst. of Technol., Hoboken, NJ, USA
Abstract :
A novel approach is discussed relative to unsupervised learning in a single-layer linear neural network. An optimality principle is proposed which is based on preserving maximal information in the output units. The unsupervised learning rule is based on a Hebbian learning rule. This learning rule finds the principal components of the input correlation matrix. For patterns classification, and image coding, the authors have modified the learning rule which finds the Boolean eigenvector
Keywords :
Boolean functions; learning systems; neural nets; pattern recognition; picture processing; Boolean eigenvector; Hebbian learning rule; image coding; input correlation matrix; neuron feature detection; optimality principle; patterns classification; single-layer linear neural network; unsupervised learning; Backpropagation; Computer vision; Eigenvalues and eigenfunctions; Hebbian theory; Image coding; Neural networks; Neurons; Pattern classification; Unsupervised learning; Vectors;
Conference_Titel :
Computers and Communications, 1989. Conference Proceedings., Eighth Annual International Phoenix Conference on
Conference_Location :
Scottsdale, AZ, USA
Print_ISBN :
0-8186-1918-x
DOI :
10.1109/PCCC.1989.37418
