VLSI design of compact and high-precision analog neural network processors

作者: J. Choi , B.J. Sheu

DOI: 10.1109/IJCNN.1992.226916

关键词:

摘要: The design of an analog VLSI neural network processor for scientific and engineering applications such as pattern recognition image compression is described. backpropagation self-organization learning schemes in artificial networks require high-precision multiplication summation. presented performs high-speed feedforward computation parallel. A digital signal or a host computer can be used updating synapse weights during the phase. computing blocks consist matrix input output neuron arrays. composed current-to-voltage converter sigmoid function generator with controllable voltage gain. An improved Gilbert multiplier design. neurons are tailored to reduce settling time minimize silicon area that implementation. >

参考文章(6)
M Banu, Y Tsividis, None, Floating voltage-controlled resistors in CMOS technology Electronics Letters. ,vol. 18, pp. 678- 679 ,(1982) , 10.1049/EL:19820461
T. Kohonen, The self-organizing map Proceedings of the IEEE. ,vol. 78, pp. 1464- 1480 ,(1990) , 10.1109/5.58325
Paul W. Hollis, John S. Harper, John J. Paulos, The effects of precision constraints in a backpropagation learning network Neural Computation. ,vol. 2, pp. 363- 373 ,(1990) , 10.1162/NECO.1990.2.3.363
Carver Mead, Analog VLSI and Neural Systems ,(1989)
A. Moore, J. Allman, R.M. Goodman, A real-time neural system for color constancy IEEE Transactions on Neural Networks. ,vol. 2, pp. 237- 247 ,(1991) , 10.1109/72.80334
C. Tomovich, MOSIS - A gateway to silicon IEEE Circuits & Devices. ,vol. 4, pp. 22- 23 ,(1988) , 10.1109/101.936