Restricting exchangeable nonparametric distributions

TL;DR

This paper introduces a new class of exchangeable nonparametric priors that restrict existing models to better control the distribution of features per data point, improving modeling flexibility.

Contribution

It proposes a novel approach to restrict the domain of exchangeable nonparametric distributions, enabling explicit control over the number of features per data point.

Findings

Improved modeling of feature counts in data sets

Enhanced performance on datasets with specific feature distributions

Flexible prior construction for exchangeable models

Abstract

Distributions over exchangeable matrices with infinitely many columns, such as the Indian buffet process, are useful in constructing nonparametric latent variable models. However, the distribution implied by such models over the number of features exhibited by each data point may be poorly- suited for many modeling tasks. In this paper, we propose a class of exchangeable nonparametric priors obtained by restricting the domain of existing models. Such models allow us to specify the distribution over the number of features per data point, and can achieve better performance on data sets where the number of features is not well-modeled by the original distribution.