An Infinite Factor Model Hierarchy via a Noisy-or Mechanism
Abstract
The Indian Buffet Process is a Bayesian nonparametric approach that models objects as arising from an infinite number of latent factors. Here we extend the latent factor model framework to two or more unbounded layers of latent factors. From a generative perspective, each layer defines a conditional \emph{factorial} prior distribution over the binary latent variables of the layer below via a noisy-or mechanism. We explore the properties of the model with two empirical studies, one digit recognition task and one music tag data experiment.
Cite
Text
Eck et al. "An Infinite Factor Model Hierarchy via a Noisy-or Mechanism." Neural Information Processing Systems, 2009.Markdown
[Eck et al. "An Infinite Factor Model Hierarchy via a Noisy-or Mechanism." Neural Information Processing Systems, 2009.](https://mlanthology.org/neurips/2009/eck2009neurips-infinite/)BibTeX
@inproceedings{eck2009neurips-infinite,
title = {{An Infinite Factor Model Hierarchy via a Noisy-or Mechanism}},
author = {Eck, Douglas and Bengio, Yoshua and Courville, Aaron C.},
booktitle = {Neural Information Processing Systems},
year = {2009},
pages = {405-413},
url = {https://mlanthology.org/neurips/2009/eck2009neurips-infinite/}
}