/MEIBP

Maximization-Expectation Indian Buffet Process (submodular IBP)

Primary LanguageMATLABMIT LicenseMIT

MEIBP

Maximization-Expectation Indian Buffet Process (submodular IBP)

This code implements the nonnegative linear-Gaussian IBP described in Scaling the Indian Buffet Process via Submodular Maximization [1]. See [2] for background material.

See meibp_test.m for example usage on a simple synthetic dataset consisting of binary latent factors with nice structure. Please email the author with any questions.

Colorado Reed; gmail address: colorado.j.reed
Department of Engineering
University of Cambridge
Cambridge, UK

This software is open-source, released under the terms of the The MIT License (see LICENSE).

[1] Scaling the Indian Buffet Process via Submodular Maximization Colorado Reed and Zoubin Ghahramani, http://arxiv.org/abs/1304.3285

[2] Infinite latent feature models and the Indian buffet process Tom Griffiths and Zoubin Ghahramani, NIPS 2006