Learning to represent spatial transformations with factored higher-order Boltzmann machines
Neural Computation, pp. 1473-1492, 2010.
interaction tensorlarger image patchimage transformationlow-rank approximationhidden unitMore(11+)
To allow the hidden units of a restricted Boltzmann machine to model the transformation between two successive images, Memisevic and Hinton (2007) introduced three-way multiplicative interactions that use the intensity of a pixel in the first image as a multiplicative gain on a learned, symmetric weight between a pixel in the second image...More
Full Text (Upload PDF)
PPT (Upload PPT)