Sparse Three-parameter Restricted Indian Buffet Process for Understanding International Trade

TL;DR

This paper introduces a flexible Bayesian nonparametric model combining three-parameter and restricted Indian buffet processes for analyzing high-dimensional count data, specifically applied to understanding countries' economic structures.

Contribution

The paper develops a novel sparse latent feature model that enhances interpretability and flexibility by integrating two Indian buffet process variants, improving analysis of economic data.

Findings

Better approximation of input distributions.

More interpretable inferred topics.

Enhanced sparsity capture in data.

Abstract

This paper presents a Bayesian nonparametric latent feature model specially suitable for exploratory analysis of high-dimensional count data. We perform a non-negative doubly sparse matrix factorization that has two main advantages: not only we are able to better approximate the row input distributions, but the inferred topics are also easier to interpret. By combining the three-parameter and restricted Indian buffet processes into a single prior, we increase the model flexibility, allowing for a full spectrum of sparse solutions in the latent space. We demonstrate the usefulness of our approach in the analysis of countries' economic structure. Compared to other approaches, empirical results show our model's ability to give easy-to-interpret information and better capture the underlying sparsity structure of data.