作者: Wesley E. Foor
DOI:
关键词:
摘要: Abstract : An adaptive optical radial basis function neural network classifier is experimentally demonstrated. We describe a spatially multiplexed system incorporating on-line adaptation of weights and widths to provide robustness imperfections noise. The computes the Euclidean distances between 100-dimensional input vector 198 stored reference patterns in parallel using dual vector-matrix multipliers contrast-reversing spatial light modulator. Software used emulate an analog electronic chip that performs learning widths. experimental recognition rate 92.7% correct out 300 testing samples achieved with training versus 31.0% for non-adaptive training. compare results detailed computer model order analyze influence various noise sources on performance. (KAR) P. 3