This approach has made variational inference applicable to alarge class ofcomplexgenerativemodels. However,manychallenges remain.

Topic models, which have the ability to uncover the hidden semantic structure in a text corpus, have been widely applied to text analysis.

We propose a scalable algorithm for model selection in sigmoid belief networks (SBNs), based on the factorized asymptotic Bayesian (FAB) framework. We derive the corresponding generalized factorized information criterion (gFIC) for the SBN, which is proven to be statistically consistent with the marginal log-likelihood. To capture the dependencies within hidden variables in SBNs, a recognition network is employed to model the variational distribution. The resulting algorithm, which we call FABIA, can simultaneously execute both model selection and inference by maximizing the lower bound of gFIC. On both synthetic and real data, our experiments suggest that FABIA, when compared to state-of-the-art algorithms for learning SBNs, $(i)$ produces a more concise model, thus enabling faster testing; $(ii)$ improves predictive performance; $(iii)$ accelerates convergence; and $(iv)$ prevents overfitting.

The factorized asymptotic Bayesian ( FAB) approach has recently been shown as a scalable model-selection framework for latent variable models.

We present a probabilistic framework for nonlinearities, based on doubly truncated Gaussian distributions.

Neural Information Processing Systems http://nips.cc/

V ariational approaches are often used to approximate intractable posteriors or normalization constants in hierarchical latent variable models.

Recently, considerable research effort has been devoted to developing deep architectures for topic models to learn topic structures.

Moreover, we develop stochastic gradient MCMC inference that is scalable to very long multivariate count time series.

Recently, considerable research effort has been devoted to developing deep architectures for topic models to learn topic structures.