Optoelectronic array that computes error and weight modification for a bipolar optical neural network.

The design, fabrication, and results of an optoelectronic device that computes the weight changes required by the delta-rule learning algorithm and encodes the result on a pair of optical beams are presented. This very-large-scale-integrated ferroelectric liquid-crystal array was designed specifically to permit bipolar optoelectronic neural networks to learn without the limitations of an external controlling computer. The device contains 64 smart pixels, which represent the processing elements in an artificial neural network. Each processing element consists of two photodetectors, a current-to-voltage converter, two comparators, two switches, and two output liquid-crystal modulating pads. The device has a measured contrast ratio of 10:1, a 10%-to-90% rise time of 350 micros, and a 90%-to-10% fall time of 150 micros.

[1]  K.M. Johnson,et al.  Liquid crystal-on-silicon smart pixel spatial light modulators , 1995, Technical Digest. CLEO/Pacific Rim'95. The Pacific Rim Conference on Lasers and Electro-Optics.

[2]  D A Jared,et al.  Electrically addressed spatial light modulator that uses a dynamic memory. , 1991, Optics letters.

[3]  M. Robinson.  Analog nonvolatile data storage and update in an optically addressed spatial light modulator. , 1992, Optics letters.

[4]  D. Brady,et al.  Adaptive optical networks using photorefractive crystals. , 1988, Applied optics.

[5]  T. J. Drabik,et al.  Silicon VLSI/ferroelectric liquid crystal technology for micropower optoelectronic computing devices. , 1990, Applied optics.

[6]  D. Jared,et al.  Optically addressed thresholding very-large-scale-integration/liquid-crystal spatial light modulators. , 1991, Optics letters.

[7]  Demetri Psaltis,et al.  Higher order associative memories and their optical implementations , 1988, Neural Networks.

[8]  Mark A. Handschy,et al.  High-speed binary optically addressed spatial light modulator , 1989 .

[9]  K M Johnson,et al.  Optical implementation of a second-order neural network. , 1991, Optics letters.

[10]  A Von Lehmen,et al.  Holographic associative memory for word-break recognition. , 1989, Optics letters.

[11]  K. Johnson,et al.  Motivations for using ferroelectric liquid crystal spatial light modulators in neurocomputing. , 1989, Applied optics.

[12]  T. J. Drabik,et al.  Ferroelectric-liquid-crystal/silicon-integrated-circuit spatial light modulator. , 1990, Optics letters.

[13]  K M Johnson,et al.  Noise analysis of polarization-based optoelectronic connectionist machines. , 1992, Applied optics.

[14]  Colin Giles,et al.  Learning, invariance, and generalization in high-order neural networks. , 1987, Applied optics.

[15]  P. Cambon,et al.  Bistability and nonlinearity in optically addressed ferroelectric liquid-crystal spatial light modulators: applications to neurocomputing. , 1992, Applied optics.

[16]  Shoichi Kurita,et al.  High‐speed light valve using an amorphous silicon photosensor and ferroelectric liquid crystals , 1987 .

[17]  K Wagner,et al.  Winner-take-all spatial light modulator. , 1992, Optics letters.

[18]  Lin Zhang,et al.  Optical connectionist machine with polarization-based bipolar weight values , 1989 .

[19]  D Z Anderson,et al.  Dynamic optical interconnects: volume holograms as optical two-port operators. , 1987, Applied optics.

[20]  M. Handschy,et al.  Ferroelectric Liquid Crystal Electro-Optics Using the Surface Stabilized Structure , 1983 .

[21]  N. Clark,et al.  Submicrosecond bistable electro‐optic switching in liquid crystals , 1980 .

[22]  D. G. Vass,et al.  Evolutionary development of advanced liquid crystal spatial light modulators. , 1989, Applied optics.

[23]  D J McKnight,et al.  Development of a spatial light modulator: a randomly addressed liquid-crystal-over-nMOS array. , 1989, Applied optics.

[24]  Neil Collings,et al.  An amorphous silicon/chiral smectic spatial light modulator , 1988 .