A subspace method for large-scale trace ratio problems

Giulia Ferrandi (Corresponding author), Michiel E. Hochstenbach, M. Rosário Oliveira

Research output: Contribution to journalArticleAcademicpeer-review

3 Downloads (Pure)

Abstract

A subspace method is introduced to solve large-scale trace ratio problems. This approach is matrix-free, requiring only the action of the two matrices involved in the trace ratio. At each iteration, a smaller trace ratio problem is addressed in the search subspace. Additionally, the algorithm is endowed with a restarting strategy, that ensures the monotonicity of the trace ratio value throughout the iterations. The behavior of the approximate solution is investigated from a theoretical viewpoint, extending existing results on Ritz values and vectors, as the angle between the search subspace and the exact solution approaches zero. Numerical experiments in multigroup classification show that this new subspace method tends to be more efficient than iterative approaches relying on (partial) eigenvalue decompositions at each step.

Original languageEnglish
Article number108108
Number of pages17
JournalComputational Statistics and Data Analysis
Volume205
DOIs
Publication statusPublished - May 2025

Bibliographical note

Publisher Copyright:
© 2024 The Author(s)

Keywords

  • Davidson's method
  • Fisher's discriminant analysis
  • Linear dimensionality reduction
  • Multigroup classification
  • Subspace method
  • Trace ratio

Fingerprint

Dive into the research topics of 'A subspace method for large-scale trace ratio problems'. Together they form a unique fingerprint.

Cite this