Efficient Training of LDA on a GPU by Mean-for-Mode Estimation

Abstract

We introduce Mean-for-Mode estimation, a variant of an uncollapsed Gibbs sampler that we use to train LDA on a GPU. The algorithm combines benefits of both uncollapsed and collapsed Gibbs samplers. Like a collapsed Gibbs sampler — and unlike an uncollapsed Gibbs sampler — it has good statistical performance, and can use sampling complexity reduction techniques such as sparsity. Meanwhile, like an uncollapsed Gibbs sampler — and unlike a collapsed Gibbs sampler — it is embarrassingly parallel, and can use approximate counters.

Cite

Text

Tristan et al. "Efficient Training of LDA on a GPU by Mean-for-Mode Estimation." International Conference on Machine Learning, 2015.

Markdown

[Tristan et al. "Efficient Training of LDA on a GPU by Mean-for-Mode Estimation." International Conference on Machine Learning, 2015.](https://mlanthology.org/icml/2015/tristan2015icml-efficient/)

BibTeX

@inproceedings{tristan2015icml-efficient,
  title     = {{Efficient Training of LDA on a GPU by Mean-for-Mode Estimation}},
  author    = {Tristan, Jean-Baptiste and Tassarotti, Joseph and Steele, Guy},
  booktitle = {International Conference on Machine Learning},
  year      = {2015},
  pages     = {59-68},
  volume    = {37},
  url       = {https://mlanthology.org/icml/2015/tristan2015icml-efficient/}
}