ACL–adaptive correction of learning parameters for backpropagation based algorithms
Publikation: Beiträge in Sammelwerken › Aufsätze in Konferenzbänden › Forschung
Standard
ACL–adaptive correction of learning parameters for backpropagation based algorithms. / Wilk, Jan; Wilk, Eva; Gobel, H.
IJCNN '99, International Joint Conference on Neural Networks: Washington, DC, July 10 - 16, 1999, Proceedings. Piscataway : IEEE - Institute of Electrical and Electronics Engineers Inc., 1999. S. 1749-1752 (International Joint Conference on Neural Networks. Proceedings; Band 3).Publikation: Beiträge in Sammelwerken › Aufsätze in Konferenzbänden › Forschung
Harvard
APA
Vancouver
Bibtex
}
RIS
TY - CHAP
T1 - ACL–adaptive correction of learning parameters for backpropagation based algorithms
AU - Wilk, Jan
AU - Wilk, Eva
AU - Gobel, H.
PY - 1999/7
Y1 - 1999/7
N2 - We present an improvement of backpropagation learning (BP) for Sigma-Pi networks with adaptive correction of the learning parameters (ACL). An improvement of convergency is achieved by using the information value, change of the output error and the validity of Funahashi's theorem to analytically determine values for the learning parameters momentum, learning rate and learning motivation in each learning step. Its application to a neural-network based approximation of continuous input-output mappings with high accuracy yields very good results: the number of training periods of ACL BP learning is smaller than the corresponding number of training periods using other BP based learning rules.
AB - We present an improvement of backpropagation learning (BP) for Sigma-Pi networks with adaptive correction of the learning parameters (ACL). An improvement of convergency is achieved by using the information value, change of the output error and the validity of Funahashi's theorem to analytically determine values for the learning parameters momentum, learning rate and learning motivation in each learning step. Its application to a neural-network based approximation of continuous input-output mappings with high accuracy yields very good results: the number of training periods of ACL BP learning is smaller than the corresponding number of training periods using other BP based learning rules.
KW - Informatics
U2 - 10.1109/IJCNN.1999.832641
DO - 10.1109/IJCNN.1999.832641
M3 - Article in conference proceedings
SN - 0-7803-5529-6
SN - 0-7803-5530-X
T3 - International Joint Conference on Neural Networks. Proceedings
SP - 1749
EP - 1752
BT - IJCNN '99, International Joint Conference on Neural Networks
PB - IEEE - Institute of Electrical and Electronics Engineers Inc.
CY - Piscataway
T2 - International Joint Conference on Neural Networks - 1999
Y2 - 10 July 1999 through 16 July 1999
ER -