First provide a summary of the paper, and then address the following criteria: Quality, clarity, originality and significance. The authors describe a two novel inference methods for the correlated topic model (CTM). They build on analytic results for the conditional logistic normal likelihood to arrive at a fast, easily parallelized exact inference. This leads to an approximate sampling method for producing Polya-Gamma variates. Finally, they propose a method for efficiently drawing samples in the presence of sparsity.

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

Logistic-normal topic models can effectively discover correlation structures among latent topics. However, their inference remains a challenge because of the non-conjugacy between the logistic-normal prior and multinomial topic mixing proportions. Existing algorithms either make restricting mean-field assumptions or are not scalable to large-scale applications. This paper presents a partially collapsed Gibbs sampling algorithm that approaches the provably correct distribution by exploring the ideas of data augmentation. To improve time efficiency, we further present a parallel implementation that can deal with large-scale applications and learn the correlation structures of thousands of topics from millions of documents. Extensive empirical results demonstrate the promise.