Rethinking PCA Through Duality

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Quan, Jan, Suykens, Johan, Patrinos, Panagiotis
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915566054277120
author Quan, Jan
Suykens, Johan
Patrinos, Panagiotis
author_facet Quan, Jan
Suykens, Johan
Patrinos, Panagiotis
contents Motivated by the recently shown connection between self-attention and (kernel) principal component analysis (PCA), we revisit the fundamentals of PCA. Using the difference-of-convex (DC) framework, we present several novel formulations and provide new theoretical insights. In particular, we show the kernelizability and out-of-sample applicability for a PCA-like family of problems. Moreover, we uncover that simultaneous iteration, which is connected to the classical QR algorithm, is an instance of the difference-of-convex algorithm (DCA), offering an optimization perspective on this longstanding method. Further, we describe new algorithms for PCA and empirically compare them with state-of-the-art methods. Lastly, we introduce a kernelizable dual formulation for a robust variant of PCA that minimizes the $l_1$ deviation of the reconstruction errors.
format Preprint
id arxiv_https___arxiv_org_abs_2510_18130
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Rethinking PCA Through Duality
Quan, Jan
Suykens, Johan
Patrinos, Panagiotis
Machine Learning
Optimization and Control
Motivated by the recently shown connection between self-attention and (kernel) principal component analysis (PCA), we revisit the fundamentals of PCA. Using the difference-of-convex (DC) framework, we present several novel formulations and provide new theoretical insights. In particular, we show the kernelizability and out-of-sample applicability for a PCA-like family of problems. Moreover, we uncover that simultaneous iteration, which is connected to the classical QR algorithm, is an instance of the difference-of-convex algorithm (DCA), offering an optimization perspective on this longstanding method. Further, we describe new algorithms for PCA and empirically compare them with state-of-the-art methods. Lastly, we introduce a kernelizable dual formulation for a robust variant of PCA that minimizes the $l_1$ deviation of the reconstruction errors.
title Rethinking PCA Through Duality
topic Machine Learning
Optimization and Control
url https://arxiv.org/abs/2510.18130