Empirical Inference Conference Paper 2004

A kernel view of the dimensionality reduction of manifolds

We interpret several well-known algorithms for dimensionality reduction of manifolds as kernel methods. Isomap, graph Laplacian eigenmap, and locally linear embedding (LLE) all utilize local neighborhood information to construct a global embedding of the manifold. We show how all three algorithms can be described as kernel PCA on specially constructed Gram matrices, and illustrate the similarities and differences between the algorithms with representative examples.

Author(s): Ham, J. and Lee, DD. and Mika, S. and Schölkopf, B.
Book Title: Proceedings of the Twenty-First International Conference on Machine Learning
Pages: 369-376
Year: 2004
Day: 0
Editors: CE Brodley
Publisher: ACM
Bibtex Type: Conference Paper (inproceedings)
Address: New York, NY, USA
Event Name: ICML 2004
Event Place: Banff, Alberta, Canada
Electronic Archiving: grant_archive
Note: also appeared as MPI-TR 110
Organization: Max-Planck-Gesellschaft
School: Biologische Kybernetik
Links:

BibTex

@inproceedings{2326,
  title = {A kernel view of the dimensionality reduction of manifolds},
  booktitle = {Proceedings of the Twenty-First International Conference on Machine Learning},
  abstract = {We interpret several well-known algorithms for dimensionality reduction of manifolds as kernel methods. Isomap, graph Laplacian eigenmap, and locally linear embedding (LLE) all utilize local neighborhood information to construct a global embedding of the manifold. We show how all three algorithms can be described as kernel PCA on specially constructed
  Gram matrices, and illustrate the similarities and differences between the algorithms with representative examples.},
  pages = {369-376},
  editors = {CE Brodley},
  publisher = {ACM},
  organization = {Max-Planck-Gesellschaft},
  school = {Biologische Kybernetik},
  address = {New York, NY, USA},
  year = {2004},
  note = {also appeared as MPI-TR 110},
  slug = {2326},
  author = {Ham, J. and Lee, DD. and Mika, S. and Sch{\"o}lkopf, B.}
}