An Optical Thresholding Perceptron

• IDIAP, 1997

An implementation of an optical perceptron with a soft optical threshold trained with an adapted BP algorithm is described as a precursor to an optical multilayer perceptron (MLP). It has 64 inputs and ten outputs. The soft threshold is implemented by a liquid crystal light valve. Experimental results on perceptron recall are also reported. The effect of a modified grey-scale to weight mapping for weight levels implemented by LCTVs is evaluated based on the results of handwritten digit recognition.

Keywords: neuron ; learning ; perry

Note:

Published in ``Proceedings of the Workshop on Optics and Computer Science''

Reference

• EPFL-REPORT-82437
• URL: http://publications.idiap.ch/downloads/reports/1997/rr97-16.pdf

Record created on 2006-03-10, modified on 2017-05-10