Learning of Highly-Filtered Data Manifold Using Spectral Methods

TitleLearning of Highly-Filtered Data Manifold Using Spectral Methods
Publication TypeConference Paper
Year of Publication2009
AuthorsRoderick, O, Safro, I
Date Published02/2009
Other NumbersANL/MCS-P1586-0209
Abstract

<p>We propose a scheme for improving existing singular value decomposition-based tools for recovering and predicting decisions. Our main contribution is an investigation of advantages of using a functional, rather than popular linear approximation of the response of an unknown, complex model. A significant attractive feature of the method is the demonstrated ability to make predictions based on a highly filtered data set.</p>

PDFhttp://www.mcs.anl.gov/papers/P1586-revised.pdf