We present a novel, scalable and Bayesian approach to modelling the occurrence of pairs (i, j) drawn from a large vocabulary. Our practical interest is in modelling (user, item) pairs in a recommender system, for which we present state of the art results on Xbox movie viewing data. The observed pairs are assumed to be generated by a simple popularity based selection process followed by censoring using a preference function. By basing inference on the well-founded principle of variational bounding, and using new site-independent bounds, we show how a scalable inference procedure can be obtained for large data sets. The model is a plausible alternative to modelling discrete densities with a bilinear softmax function.
|Number of pages||8|
|State||Published - 2014|
|Event||NEURAL INFORMATION PROCESSING SYSTEMS WORKSHOP, NIPS 2014 - Montreal, Quebec, Canada|
Duration: 8 Dec 2014 → 13 Dec 2014
|Workshop||NEURAL INFORMATION PROCESSING SYSTEMS WORKSHOP, NIPS 2014|
|Abbreviated title||NIPS 2014|
|Period||8/12/14 → 13/12/14|