Scalable Bayesian Non-Negative Tensor Factorization for Massive Count Data
We present a Bayesian non-negative tensor factorization model for count-valued tensor data, and develop scalable inference algorithms (both batch and online) for dealing with massive tensors. Our generative model can handle overdispersed counts as well as infer the rank of the decomposition. Moreover, leveraging a reparameterization of the Poisson distribution as a multinomial facilitates conjugacy in the model and enables simple Gibbs sampling and variational Bayes (VB) inference updates. We also develop a set of online inference algorithms that allow scaling up the model to massive tensors. We apply our framework on diverse real-world applications, such as analyzing a political science database, multiway topic modeling on a scientific publications database, and analyzing household transactions data.
Tuesday, September 8, 2015 - 15:25 to 15:50