A Learning Algorithm for a Neural Network LSI with Restricted Integer Weights

Tomohisa KIMURA  Takeshi SHIMA  

IEICE TRANSACTIONS on Electronics   Vol.E80-C   No.7   pp.983-989
Publication Date: 1997/07/25
Online ISSN: 
Print ISSN: 0916-8516
Type of Manuscript: Special Section PAPER (Special Issue on New Concept Device and Novel Architecture LSIs)
Category: Neural Networks and Chips
neural networks,  learning,  analog,  LSI,  accuracy,  

Full Text: PDF(536.4KB)>>
Buy this Article

A novel learning algorithm for a neural network LSI which has low resolution synapse weights is proposed. Following a brief discussion of the synapse weight adaptation mechanism in the gradient descent scheme, we propose a way of achieving relaxation from the influence of discretized weight. Restriction of the number of synapses to be updated in one learning iteration is effective to relax the influence. Simulation results support the effectiveness of this learning algorithm. Low resolution synapses will be practical to realize large-scale neural network LSIs.