English

Non-exchangeable feature allocation models with sublinear growth of the feature sizes

Machine Learning 2020-03-31 v1 Machine Learning

Abstract

Feature allocation models are popular models used in different applications such as unsupervised learning or network modeling. In particular, the Indian buffet process is a flexible and simple one-parameter feature allocation model where the number of features grows unboundedly with the number of objects. The Indian buffet process, like most feature allocation models, satisfies a symmetry property of exchangeability: the distribution is invariant under permutation of the objects. While this property is desirable in some cases, it has some strong implications. Importantly, the number of objects sharing a particular feature grows linearly with the number of objects. In this article, we describe a class of non-exchangeable feature allocation models where the number of objects sharing a given feature grows sublinearly, where the rate can be controlled by a tuning parameter. We derive the asymptotic properties of the model, and show that such model provides a better fit and better predictive performances on various datasets.

Keywords

Cite

@article{arxiv.2003.13491,
  title  = {Non-exchangeable feature allocation models with sublinear growth of the feature sizes},
  author = {Giuseppe Di Benedetto and François Caron and Yee Whye Teh},
  journal= {arXiv preprint arXiv:2003.13491},
  year   = {2020}
}

Comments

Accepted to AISTATS 2020

R2 v1 2026-06-23T14:32:01.659Z