DocumentCode
295976
Title
An empirical study of the time complexity of various error functions with conjugate gradient backpropagation
Author
Lister, Raymond ; Stone, James V.
Author_Institution
Sch. of Comput. Sci., Queensland Univ. of Technol., Brisbane, Qld., Australia
Volume
1
fYear
1995
fDate
Nov/Dec 1995
Firstpage
237
Abstract
Describes an empirical comparison of the scaling behaviour of six error functions, on a conjugate gradient form of backpropagation. The authors classify the functions according to the limit behaviours of their respective error signals, as the target value and the actual output value approach opposite extremes. These limit behaviours are zero limit, finite limit, and infinite limit. Despite such a wide divergence in their limit behaviours, the authors find that all six error functions exhibit a median run-time order of approximately O(N4) on the N-2-N encoder. This result indicates that, while some factors affecting the scaling behaviour of standard and conjugate gradient backpropagation have been previously identified (such as saturation), other factors remain unidentified
Keywords
backpropagation; computational complexity; conjugate gradient methods; multilayer perceptrons; conjugate gradient backpropagation; error functions; limit behaviours; median run-time order; scaling behaviour; time complexity; Australia; Backpropagation; Computer hacking; Counting circuits; Error correction; H infinity control; Runtime; Shape; Taxonomy; Time of arrival estimation;
fLanguage
English
Publisher
ieee
Conference_Titel
Neural Networks, 1995. Proceedings., IEEE International Conference on
Conference_Location
Perth, WA
Print_ISBN
0-7803-2768-3
Type
conf
DOI
10.1109/ICNN.1995.488101
Filename
488101
Link To Document