Back to Results
First PageMeta Content
Mathematics / Recurrent neural network / Artificial neural network / Feedforward neural network / Activation function / Backpropagation / Sigmoid function / Autoencoder / Logistic function / Neural networks / Cybernetics / Science


Understanding the difficulty of training deep feedforward neural networks Xavier Glorot Yoshua Bengio DIRO, Universit´e de Montr´eal, Montr´eal, Qu´ebec, Canada
Add to Reading List

Document Date: 2010-03-31 18:50:33


Open Document

File Size: 1,57 MB

Share Result on Facebook

City

Courville / New York / /

Company

Solla S. A. / Neural Networks / MIT Press / Weston / /

Country

United States / Italy / /

Facility

Carnegie Mellon University / University of Toronto / The Robotics Institute / /

IndustryTerm

softsign networks / hyperbolic tangent network / hyperbolic tangent networks / investigative tool / online learning scenario / Online Learning / energy-based model / deep dense networks / deep networks / online training / natural language processing / deeper networks / softsign network / initialized network / online setting / sigmoid networks / tanh network / tanh networks / learning algorithm / deep feedforward networks / /

MarketIndex

set 300 / /

Organization

Robotics Institute / MIT / Carnegie Mellon University / University of Toronto / Pattern Analysis and Machine Intelligence / /

Person

Xavier Glorot Yoshua Bengio / /

ProvinceOrState

Sardinia / /

PublishedMedium

IEEE Transactions on Pattern Analysis and Machine Intelligence / Machine Learning / Complex Systems / The Journal of Machine Learning Research / /

Technology

natural language processing / neural network / artificial intelligence / Machine Learning / /

URL

http /

SocialTag