Matthias Seeger
We propose a simple taxonomy of probabilistic graphical models for the semi-supervised learning problem. We give some broad classes of algorithms for each of the families and point to specific realizations in the literature. Finally, we shed more detailed ...
MIT Press2006