FastEx: Hash Clustering with Exponential Families

Abstract

Clustering is a key component in data analysis toolbox. Despite its importance, scalable algorithms often eschew rich statistical models in favor of simpler descriptions such as $k$-means clustering. In this paper we present a sampler, capable of estimating mixtures of exponential families. At its heart lies a novel proposal distribution using random projections to achieve high throughput in generating proposals, which is crucial for clustering models with large numbers of clusters.

Cite

Text

Ahmed et al. "FastEx: Hash Clustering with Exponential Families." Neural Information Processing Systems, 2012.

Markdown

[Ahmed et al. "FastEx: Hash Clustering with Exponential Families." Neural Information Processing Systems, 2012.](https://mlanthology.org/neurips/2012/ahmed2012neurips-fastex/)

BibTeX

@inproceedings{ahmed2012neurips-fastex,
  title     = {{FastEx: Hash Clustering with Exponential Families}},
  author    = {Ahmed, Amr and Ravi, Sujith and Smola, Alex J. and Narayanamurthy, Shravan M.},
  booktitle = {Neural Information Processing Systems},
  year      = {2012},
  pages     = {2798-2806},
  url       = {https://mlanthology.org/neurips/2012/ahmed2012neurips-fastex/}
}