OPTICAL IMPLEMENTATION OF A 2ND-ORDER NEURAL NETWORK

被引:19
作者
ZHANG, L
ROBINSON, MG
JOHNSON, KM
机构
[1] Center for Optoelectronic Computing Systems, Department of Electrical and Computer Engineering, University of Colorado at Boulder, Boulder, CO
关键词
D O I
10.1364/OL.16.000045
中图分类号
O43 [光学];
学科分类号
070207 ; 0803 ;
摘要
An optical implementation of a single-layer, second-order neural network is presented. The quadratic products are obtained by passing the optical beam twice through the input ferroelectric liquid-crystal (FLC) spatial light modulator (SLM), with the interconnection weights being implemented by a further two-dimensional 128 x 128 FLC SLM. The machine successfully associates eight randomly chosen pattern-target pairs (dimensions 16 and 4, respectively) and can learn the parity association. Translation invariance is also demonstrated. Results from a computer model indicate that input SLM contrast ratios of 4:1 and electronic noise of 10% of the maximum output can be tolerated.
引用
收藏
页码:45 / 47
页数:3
相关论文
共 6 条
  • [1] LEARNING, INVARIANCE, AND GENERALIZATION IN HIGH-ORDER NEURAL NETWORKS
    GILES, CL
    MAXWELL, T
    [J]. APPLIED OPTICS, 1987, 26 (23): : 4972 - 4978
  • [2] HORAN P, 1990, 1990 P INT TOP M OPT, P102
  • [3] OPTICAL CONNECTIONIST MACHINE WITH POLARIZATION-BASED BIPOLAR WEIGHT VALUES
    KRANZDORF, M
    BIGNER, BJ
    ZHANG, L
    JOHNSON, KM
    [J]. OPTICAL ENGINEERING, 1989, 28 (08) : 844 - 848
  • [4] REID MB, 1989, P INT JOINT C NEURAL
  • [5] ROBINSON MG, 1990, P SOC PHOTO-OPT INS, V1215, P389, DOI 10.1117/12.18086
  • [6] VONLEHMEN A, 1980, OPT LETT, V5, P279