Title :
DRBP: dynamically reinforced BP-based ANN-training
Author :
Cheng, Xiang S. ; Backer, Eric ; Gerbrands, Jan J.
Author_Institution :
Philips CFT, Eindhoven, Netherlands
fDate :
30 Aug-3 Sep 1992
Abstract :
Describes a new training method, the DRBP-algorithm, for sigmoid-function based multilayer networks. The key step in DRBP is the dynamical selection and autonomous control of the learning rate. Various experiments have shown that the DRBP-algorithm has achieved its goal of fast speed, secure stability and easy parameter selection in practice
Keywords :
learning systems; neural nets; optimisation; DRBP-algorithm; dynamically reinforced backpropagation training; learning systems; multilayer networks; neural nets; parameter selection; sigmoid-function; stability; Algorithm design and analysis; Artificial neural networks; Computer vision; Feedforward systems; Laboratories; Machine vision; Multi-layer neural network; Neural networks; Stability; Training data;
Conference_Titel :
Pattern Recognition, 1992. Vol.II. Conference B: Pattern Recognition Methodology and Systems, Proceedings., 11th IAPR International Conference on
Conference_Location :
The Hague
Print_ISBN :
0-8186-2915-0
DOI :
10.1109/ICPR.1992.201710
