The Stability of Kernel Principal Components Analysis and Its Relation to the Process Eigenspectrum

Christopher Williams, John S. Shawe-taylor

NeurIPS 2002 pp. 383-390

/neurips/2002/williams2002neurips-stability/

Abstract

In this paper we analyze the relationships between the eigenvalues of the m x m Gram matrix K for a kernel k(·, .) corresponding to a sample Xl, ... ,Xm drawn from a density p(x) and the eigenvalues of the corresponding continuous eigenproblem. We bound the dif(cid:173) ferences between the two spectra and provide a performance bound on kernel peA.

PDF NeurIPS Semantic Scholar

Cite

Text

Williams and Shawe-taylor. "The Stability of Kernel Principal Components Analysis and Its Relation to the Process Eigenspectrum." Neural Information Processing Systems, 2002.

Markdown

[Williams and Shawe-taylor. "The Stability of Kernel Principal Components Analysis and Its Relation to the Process Eigenspectrum." Neural Information Processing Systems, 2002.](https://mlanthology.org/neurips/2002/williams2002neurips-stability/)

BibTeX

@inproceedings{williams2002neurips-stability,
  title     = {{The Stability of Kernel Principal Components Analysis and Its Relation to the Process Eigenspectrum}},
  author    = {Williams, Christopher and Shawe-taylor, John S.},
  booktitle = {Neural Information Processing Systems},
  year      = {2002},
  pages     = {383-390},
  url       = {https://mlanthology.org/neurips/2002/williams2002neurips-stability/}
}