Variational EM algorithms for non-Gaussian latent variable models
Jason A. Palmer, Kenneth Kreutz-Delgado, D.R. Wipf, B.D. Rao · 2006
We consider criteria for variational representations of non-Gaussian la-tent variables, and derive variational EM algorithms in general form. We establish a general equivalence among convex bounding methods, evi-dence based methods, and ensemble learning/Variational Bayes methods, which has previously been demonstrated only for particular cases. 1