Statistical Convergence of Kernel CCA

Abstract

While kernel canonical correlation analysis (kernel CCA) has been applied in many problems, the asymptotic convergence of the functions estimated from a finite sample to the true functions has not yet been established. This paper gives a rigorous proof of the statistical convergence of kernel CCA and a related method (NOCCO), which provides a theoretical justification for these methods. The result also gives a sufficient condition on the decay of the regularization coefficient in the methods to ensure convergence.

Cite

Text

Fukumizu et al. "Statistical Convergence of Kernel CCA." Neural Information Processing Systems, 2005.

Markdown

[Fukumizu et al. "Statistical Convergence of Kernel CCA." Neural Information Processing Systems, 2005.](https://mlanthology.org/neurips/2005/fukumizu2005neurips-statistical/)

BibTeX

@inproceedings{fukumizu2005neurips-statistical,
  title     = {{Statistical Convergence of Kernel CCA}},
  author    = {Fukumizu, Kenji and Gretton, Arthur and Bach, Francis R.},
  booktitle = {Neural Information Processing Systems},
  year      = {2005},
  pages     = {387-394},
  url       = {https://mlanthology.org/neurips/2005/fukumizu2005neurips-statistical/}
}