Mixtures of robust probabilistic principal component analyzers
(2007) European Symposium on Artificial Neural Networks (ESANN 2007) — Location: Bruges (Belgium) (25.April.2007)
Files
MixturesofRobustProbabilisticPrincipalComponentAnalyzers.pdf
Restricted Access - Adobe PDF
- 7.15 MB
Details
- Authors
- Abstract
- Discovering low-dimensional (nonlinear) manifolds is an important problem in Machine Learning. In many applications, the data are in a high dimensional space. This can be problematic in practice due to the curse of dimensionality. Fortunately, the core of the data lies often on one or several low-dimensional manifolds. A way to handle these is to pre-process the data by nonlinear data projection techniques. Another approach is to combine local linear models. In particular, mixtures of probabilistic principal component analyzers are very attractive as each component is specifically designed to extract the local principal orientations in the data. However, an important issue is the model sensitivity to data lying off the manifold, possibly leading to mismatches between successive local models. The mixtures of robust probabilistic principal component analyzers introduced in this paper heal this problem as each component is able to cope with atypical data while identifying the local principal directions. Moreover, the standard mixture of Gaussians is a particular instance of this more general model.
- Affiliations
Citations
Archambeau, C., Delannay, N., & Verleysen, M. (2007). Mixtures of robust probabilistic principal component analyzers. Proceedings of the 2007 European Symposium on Artificial Neural Networks (ESANN 2007), p. 229-234. https://hdl.handle.net/2078.5/253909