Skip to Main content Skip to Navigation
Journal articles

LDA via L1-PCA of Whitened Data

Abstract : Principal component analysis (PCA) and Fisher's linear discriminant analysis (LDA) are widespread techniques in data analysis and pattern recognition. Recently, the L1-norm has been proposed as an alternative criterion to classical L2-norm in PCA, drawing considerable research interest on account of its increased robustness to outliers. The present work proves that, combined with a whitening preprocessing step, L1-PCA can perform LDA in an unsupervised manner, i.e., sparing the need for labelled data. Rigorous proof is given in the case of data drawn from a mixture of Gaussians. A number of numerical experiments on synthetic as well as real data confirm the theoretical findings.
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-02486881
Contributor : Vicente Zarzoso <>
Submitted on : Friday, February 21, 2020 - 11:42:10 AM
Last modification on : Thursday, March 5, 2020 - 12:20:25 PM

Identifiers

Collections

Citation

Rubén Martín-Clemente, Vicente Zarzoso. LDA via L1-PCA of Whitened Data. IEEE Transactions on Signal Processing, Institute of Electrical and Electronics Engineers, 2020, 68, pp.225-240. ⟨10.1109/TSP.2019.2955860⟩. ⟨hal-02486881⟩

Share

Metrics

Record views

28