On the training of limited precision multi-layer perceptrons

Author

Xie, Yun ; Jabri, Marwan A.

Author_Institution

Dept. of Electr. Eng., Sydney Univ., NSW, Australia

Volume

3

fYear

1992

fDate

7-11 Jun 1992

Firstpage

942

Abstract

The effects of quantization on the training dynamics of a real-valued feedforward multilayer neural network when implemented in digital hardware are analyzed. It is shown that special techniques have to be employed to train such networks where all the variables are represented by limited numbers of bits in fixed point format. A training algorithm based on the analysis called the combined search algorithm is proposed. The combined search algorithm consists of two kinds of search techniques and is easy to implement in hardware. Using intracardiac electrograms and sonar reflection pattern recognition, extensive computer simulations were conducted. The simulation results are given

Keywords

digital simulation; feedforward neural nets; learning (artificial intelligence); pattern recognition; combined search algorithm; computer simulations; digital hardware; fixed point format; intracardiac electrograms; limited precision multilayer perceptrons; quantization; real-valued feedforward multilayer neural network; sonar reflection pattern recognition; training; training dynamics; Algorithm design and analysis; Feedforward neural networks; Multi-layer neural network; Multilayer perceptrons; Neural network hardware; Neural networks; Pattern recognition; Quantization; Reflection; Sonar;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks, 1992. IJCNN., International Joint Conference on

Conference_Location

Baltimore, MD

Print_ISBN

0-7803-0559-0

Type

conf

DOI

10.1109/IJCNN.1992.227078

Filename

227078