Revisiting natural gradient for deep networks Yoshua Bengio Universit´e de Montr´eal Montr´eal QC H3C 3J7 Canada

First Page		Document Content
Date: 2014-02-18 01:51:44 Numerical analysis Mathematical analysis Mathematics Estimation theory Numerical linear algebra M-estimators Stochastic optimization Conjugate gradient method Nonlinear conjugate gradient method Gradient descent Gradient method Hessian matrix		Revisiting natural gradient for deep networks Yoshua Bengio Universit´e de Montr´eal Montr´eal QC H3C 3J7 Canada Add to Reading List Source URL: arxiv.org Download Document from Source Website File Size: 669,56 KB Share Document on Facebook

	doi:ipiInverse Problems and Imaging Volume 10, No. 1, 2016, 195–225 PRECONDITIONED CONJUGATE GRADIENT METHOD FOR DocID: 1rsPA - View Document
	The Condition of a System of Linear Equations: Alternative Derivation Roland Angst Computer Vision and Geometry Lab, Eidgenössische Technische Hochschule Zürich Zürich, Switzerland DocID: 1rcA7 - View Document
	Revisiting natural gradient for deep networks Yoshua Bengio Universit´e de Montr´eal Montr´eal QC H3C 3J7 Canada DocID: 1qzJK - View Document
	Spectral Graph Theory and its Applications Lecture 18 Solving Linear Equations November 9, 2004 DocID: 1qtGA - View Document
	MITSUBISHI ELECTRIC RESEARCH LABORATORIES http://www.merl.com Accelerated graph-based nonlinear denoising filters Knyazev, A.; Malyshev, A. TR2016-068 DocID: 1qqXv - View Document