Nonlinear dimensionality reduction for discriminative analytics of multiple datasets

Research output: Contribution to journalArticle

3 Scopus citations


Principal component analysis (PCA) is widely used for feature extraction and dimensionality reduction, with documented merits in diverse tasks involving high-dimensional data. PCA copes with one dataset at a time, but it is challenged when it comes to analyzing multiple datasets jointly. In certain data science settings however, one is often interested in extracting the most discriminative information from one dataset of particular interest (a.k.a. target data) relative to the other(s) (a.k.a. background data). To this end, this paper puts forth a novel approach, termed discriminative (d) PCA, for such discriminative analytics of multiple datasets. Under certain conditions, dPCA is proved to be least-squares optimal in recovering the latent subspace vector unique to the target data relative to background data. To account for nonlinear data correlations, (linear) dPCA models for one or multiple background datasets are generalized through kernel-based learning. Interestingly, all dPCA variants admit an analytical solution obtainable with a single (generalized) eigenvalue decomposition. Finally, substantial dimensionality reduction tests using synthetic and real datasets are provided to corroborate the merits of the proposed methods.

Original languageEnglish (US)
Article number8565879
Pages (from-to)740-752
Number of pages13
JournalIEEE Transactions on Signal Processing
Issue number3
StatePublished - Feb 1 2019



  • Principal component analysis
  • discriminative analytics
  • kernel learning
  • multiple background datasets

Cite this