PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Algorithms and literate programs for weighted low-rank approximation with missing data
I Markovsky
(2009) Technical Report. eprints, Southampton, UK.

Abstract

Linear models identification from data with missing values is posed as a weighted low-rank approximation problem with weights related to the missing values equal to zero. Alternating projections and variable projections methods for solving the resulting problem are outlined and implemented in a literate programming style, using Matlab/Octave's scripting language. The methods are evaluated on synthetic data and real data from the MovieLens data sets.

EPrint Type:Monograph (Technical Report)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Learning/Statistics & Optimisation
Theory & Algorithms
ID Code:5720
Deposited By:Ivan Markovsky
Deposited On:08 March 2010