Modeling the Manifolds of Images of Handwritten Digits
Geoffrey E. Hinton, Peter Dayan
and Michael Revow
Department of Computer Science
University of Toronto
Abstract
This paper describes two new methods for modeling the manifolds of
digitised images of handwritten digits. The models allow a priori information about
the structure of the manifolds to be combined with empirical data. Accurate modeling of
the manifolds allows digits to be discriminated using the relative probability densities
under the alternative models. One of the methods is grounded in principal components
analysis, the other in factor analysis. Both methods are based on locally linear,
low-dimensional approximations to the underlying data manifold. Links with other methods
that model the manifold are discussed.
Download [ps] [ps.gz] [pdf]
IEEE Transactions on Neural Networks, 8 65-74
[home page] [publications]