| Open Document File Size: 199,19 KBShare Result on Facebook
City New York / Paris / Prague / Granada / Lyon / Clearwater Beach / San Juan / / Company Neural Information Processing Systems / Neural Networks / MIT Press / GPU / C. Plahl A. E. / Morgan Kaufmann Publishers Inc. / OSEO / / Country Germany / Puerto Rico / France / United States / Spain / / Currency SGD / / Facility RWTH Aachen University / / IndustryTerm layer network / tandem systems / improved optimization algorithms / optimization algorithm / batch algorithms / distributed deep networks / large networks / stochastic algorithms / proposed optimization algorithm / deep belief networks / optimization algorithms / stochastic second-order optimization algorithm / automatic speech recognition systems / stochastic optimization algorithm / stochastic second-order algorithm / / Organization MIT / Computer Science Department / RWTH Aachen University / Aachen / European Union / French State / / Person A. Ng / K. Yang / G. Heigold / Ralf Schl / Bt / B. Kingsbury / V / M. Ranzato / P. Tucker / / Position final model for recognition / representative / senior chair / / Product Newbob / Newbob+ / / ProvinceOrState Ohio / New York / Florida / / PublishedMedium Machine Learning / Journal of Machine Learning Research / / Technology stochastic second-order optimization algorithm / speech recognition / machine learning / speech recognition system / 2 1 Human Language Technology / batch algorithms / optimization algorithm / Rprop algorithm / neural network / proposed algorithm / stochastic second-order algorithm / second-order algorithm / optimization algorithms / proposed optimization algorithm / improved optimization algorithms / second-order stochastic optimization algorithm / Hessian-Free algorithm / / URL http /
SocialTag |
MEAN-NORMALIZED STOCHASTIC GRADIENT FOR LARGE-SCALE DEEP LEARNING Simon Wiesler1 , Alexander Richard1 , Ralf Schl¨uter1 , Hermann Ney1,2 1 Human Language Technology and Pattern Recognition, Computer Science Department,