Recognizing Handwritten Digits Using Mixtures of Linear Models
Geoffrey E Hinton, Michael Revow and Peter Dayan
Department of Computer Science
University of Toronto
We construct a mixture of locally linear generative models of a
collection of pixel-based images of digits, and use them for recognition. Different models
of a given digit are used to capture different styles of writing, and new images are
classified by evaluating their log-likelihoods under each model. We use an EM-based
algorithm in which the M-step is computationally straightforward principal components
analysis (PCA). Incorporating tangent-plane information about expected local deformations
only requires adding tangent vectors into the sample covariance matrices for the PCA, and
it demonstrably improves performance.
Download [ps] [ps.gz] [pdf]
Advances in Neural Information Processing Systems 7. G. Tesauro,
D. S. Touretzky and T. K. Leen (Eds), pp 1015-1022 MIT Press, Cambridge MA.
[home page] [publications]