ML Anthology

Robust Probabilistic Projections

Cédric Archambeau, Nicolas Delannay, Michel Verleysen
ICML 2006 pp. 33-40
doi:10.1145/1143844.1143849 /icml/2006/archambeau2006icml-robust/

Abstract

Principal components and canonical correlations are at the root of many exploratory data mining techniques and provide standard pre-processing tools in machine learning. Lately, probabilistic reformulations of these methods have been proposed (Roweis, 1998; Tipping & Bishop, 1999b; Bach & Jordan, 2005). They are based on a Gaussian density model and are therefore, like their non-probabilistic counterpart, very sensitive to atypical observations. In this paper, we introduce robust probabilistic principal component analysis and robust probabilistic canonical correlation analysis. Both are based on a Student-t density model. The resulting probabilistic reformulations are more suitable in practice as they handle outliers in a natural way. We compute maximum likelihood estimates of the parameters by means of the EM algorithm.

PDF ICML Semantic Scholar

Cite

Text

Archambeau et al. "Robust Probabilistic Projections." International Conference on Machine Learning, 2006. doi:10.1145/1143844.1143849

Markdown

[Archambeau et al. "Robust Probabilistic Projections." International Conference on Machine Learning, 2006.](https://mlanthology.org/icml/2006/archambeau2006icml-robust/) doi:10.1145/1143844.1143849

BibTeX

@inproceedings{archambeau2006icml-robust,
  title     = {{Robust Probabilistic Projections}},
  author    = {Archambeau, Cédric and Delannay, Nicolas and Verleysen, Michel},
  booktitle = {International Conference on Machine Learning},
  year      = {2006},
  pages     = {33-40},
  doi       = {10.1145/1143844.1143849},
  url       = {https://mlanthology.org/icml/2006/archambeau2006icml-robust/}
}