NIPS
2003
13 years 6 months ago
2003
In this paper we introduce a new underlying probabilistic model for principal component analysis (PCA). Our formulation interprets PCA as a particular Gaussian process prior on a ...