Title :
The effects of quantization on multilayer neural networks
Author :
Dündar, Günhan ; Rose, Kenneth
Author_Institution :
Dept. of Electr. & Electron. Eng., Bogazici Univ., Istanbul, Turkey
fDate :
11/1/1995 12:00:00 AM
Abstract :
The effect of weight quantization in multilayer neural networks is discussed. A method is derived by which one can predict the performance degradation at the output given the properties of the network and number of bits of quantization. Predictions from this method are evaluated against simulation results. An algorithm to decrease the noise at the output is presented and the results are compared with those above
Keywords :
backpropagation; multilayer perceptrons; probability; quantisation (signal); multilayer neural networks; noise; performance degradation; weight quantization; Circuit noise; Circuit simulation; Degradation; EPROM; Multi-layer neural network; Neural networks; Neurons; Predictive models; Quantization; Switched capacitor circuits;
Journal_Title :
Neural Networks, IEEE Transactions on
