Geneva / Houston / Southampton / Wallingford / Paris / Pittsburgh / Bates / Cambridge / Laval / Médecine / College Park / Grenoble / San Juan / /
Company
Neural Networks / MIT Press / Nasr A. E. / Tokar A. S. / /
Country
France / Canada / United Kingdom / /
Currency
pence / USD / / /
Event
FDA Phase / /
Facility
Catchment Area / Trans-Canada Highway / /
IndustryTerm
experimental protocol / easier tools / inner product / distributed processing / back-propagation algorithm / water resources applications / considerable computing time / hydrologic applications / radial basis function networks / supervised training algorithm / genetic algorithm / back-propagation training algorithm / less computing time / genetic algorithms / optimization tool / nonlinear systems / feedforward networks / backpropagation algorithms / smoother network / backpropagation networks / modified genetic algorithm / /
NaturalFeature
Serein Rivers / San Juan River / Study Daily stream / Pacific Coast / /
Organization
Institut National Polytechnique de Grenoble / Carnegie Mellon Univ. / MIT / Institute for Systems Research / Center for Computational Research / Harvard Univ. / School of Computer Science / Canadian Meteorological Centre / World Meteorological Organization / US Federal Reserve / Univ. of Maryland / Natural Science and Engineering Research Council of Canada / /
Person
Van Cauwenberghe / Max Mean / Eli / Min Streamflow / /
Position
D. J. / Researcher / Environment Canada / corresponding author / R. N. / Author / model development process / Professor / feed-forward / /
Product
MLPs / /
ProgrammingLanguage
MATLAB / L / /
ProvinceOrState
Quebec / Maryland / British Columbia / Massachusetts / /
PublishedMedium
the Catchments under Study Daily / the Journal of Hydrologic Engineering / JOURNAL OF HYDROLOGIC ENGINEERING / /
Comparing Sigmoid Transfer Functions for Neural Network Multistep Ahead Streamflow Forecasting H. Yonaba1; F. Anctil2; and V. Fortin3 Abstract: One of the main problems of neural networks is the lack of consensus on how