Abstract
Optical implementations of neural networks can combine advantages of neural network adaptive parallel processing and optical free-space connectivity. Binary valued Backpropagation1, a supervised learning algorithm related to standard Backpropagation2, significantly reduces interconnection storage and computation requirements. This implementation of binary valued Backpropagation used optical matrix-vector multiplication3 to represent the forward information flow between network layers. Previous analog optical network memory systems have been described4.
© 1991 Optical Society of America
PDF ArticleMore Like This
K.J. Weible, N. Collings, W. Xue, G. Pedrini, and R. Dändliker
ME9 Optical Computing (IP) 1991
Satoshi Ishihara, Nobuyuki Kasama, Masahiko Mori, Yoshio Hayasaki, and Toyohiko Yatagai
PdP2 Optical Computing (IP) 1991
Alex V. Huynh, John F. Walkup, and Thomas F. Krile
MII8 OSA Annual Meeting (FIO) 1991