Learning Deep Sigmoid Belief Networks with Data Augmentation
Abstract
Deep directed generative models are developed. The multi-layered model is designed by stacking sigmoid belief networks, with sparsity-encouraging priors placed on the model parameters. Learning and inference of layer-wise model parameters are implemented in a Bayesian setting. By exploring the idea of data augmentation and introducing auxiliary Polya-Gamma variables, simple and efficient Gibbs sampling and mean-field variational Bayes (VB) inference are implemented. To address large-scale datasets, an online version of VB is also developed. Experimental results are presented for three publicly available datasets: MNIST, Caltech 101 Silhouettes and OCR letters.
Cite
Text
Gan et al. "Learning Deep Sigmoid Belief Networks with Data Augmentation." International Conference on Artificial Intelligence and Statistics, 2015.Markdown
[Gan et al. "Learning Deep Sigmoid Belief Networks with Data Augmentation." International Conference on Artificial Intelligence and Statistics, 2015.](https://mlanthology.org/aistats/2015/gan2015aistats-learning/)BibTeX
@inproceedings{gan2015aistats-learning,
title = {{Learning Deep Sigmoid Belief Networks with Data Augmentation}},
author = {Gan, Zhe and Henao, Ricardo and Carlson, David E. and Carin, Lawrence},
booktitle = {International Conference on Artificial Intelligence and Statistics},
year = {2015},
url = {https://mlanthology.org/aistats/2015/gan2015aistats-learning/}
}