Projected Gradient Methods for Non-negative Matrix Factorization Chih-Jen Lin

First Page		Document Content
Date: 2011-04-07 10:47:46 Multivariate statistics Operations research Linear algebra Matrix theory Non-negative matrix factorization Mathematical optimization Coordinate descent Gradient descent Least squares Numerical analysis Mathematics Statistics		Projected Gradient Methods for Non-negative Matrix Factorization Chih-Jen Lin Add to Reading List Source URL: www.csie.ntu.edu.tw Download Document from Source Website File Size: 372,93 KB Share Document on Facebook

	! Ecole Doctorale des Sciences Fondamentales Title of the thesis: Resolution of non-linear systems using Least-squares type methods. Supervisor : Arnaud Münch Laboratory : UMR 6620 – Laboratoire de Mathématiques - C DocID: 1xW2L - View Document
	July 31, 2002 To appear in J. of DEDS LEAST SQUARES POLICY EVALUATION ALGORITHMS WITH LINEAR FUNCTION APPROXIMATION1 DocID: 1vp3G - View Document
	Neurocomputing, vol.80, pp, Importance-Weighted Least-Squares Probabilistic Classifier for Covariate Shift Adaptation with DocID: 1vnb3 - View Document
	Least-squares migration using complex wavelets Mark Miller* and Nick Kingsbury, Cambridge University; Richard Hobbs, Durham University Summary The use of multi-scale decompositions has lead to advances in representation DocID: 1viXU - View Document
	FILTRANE, a Fortran 95 Filter-Trust-Region Package for Solving Nonlinear Least-Squares and Nonlinear Feasibility Problems NICHOLAS I. M. GOULD Rutherford Appleton Laboratory and DocID: 1uV8s - View Document