| Document Date: 2011-05-13 07:45:13
Open Document File Size: 294,67 KBShare Result on Facebook
City TORONTO / / Company Neural Information Processing Systems / Neural Networks / Multidimensional Recurrent Neural Networks / Learning Recurrent Neural Networks / Google / Schmidhuber (1997) Learning Recurrent Neural Networks / Dynamical Recurrent Neural Networks / Recurrent Neural Networks / / Country United States / / Event Product Issues / Product Recall / / Facility Free Optimization James Martens Ilya Sutskever University of Toronto / / IndustryTerm Dynamic bayesian networks / matrix-vector product algorithm / curvature matrix-vector product / optimization algorithms / partial solution / linear conjugate gradient algorithm / usual curvature matrix-vector product / non-zero inner product / semi-online approach / search directions / biological and artificial systems / curvature-matrix products / chaotic systems / curvature matrix-vector products / wireless communication / usual algorithm / energy / / MarketIndex set 1000 / / Organization UC Berkeley / Universitat / University of Toronto / / Person G.E. Hinton / James Martens / D.E. Rumelhart / Richard Zemel / Geoffrey Hinton / R.J. Williams / Markov Model / / Position author / / Product δn / Gauss-Newton / LSTMs / gradient descent / / ProgrammingLanguage L / / PublishedMedium Machine Learning / the Echo / / Technology speech recognition / Neural Network / matrix-vector product algorithm / linear conjugate gradient algorithm / 2nd-order optimization algorithms / HF algorithm / Machine Learning / usual algorithm / MIDI / /
SocialTag |
Learning Recurrent Neural Networks with Hessian-Free Optimization James Martens Ilya Sutskever University of Toronto, Canada