Fast nonlinear dimensionality reduction with topology preserving networks

Abstract : We present a fast alternative for the Isomap algorithm. A set of quantizers is fit to the data and a neighborhood structure based on the competitive Hebbian rule is imposed on it. This structure is used to obtain low-dimensional description of the data by means of computing geodesic distances and multi dimensional scaling. The quantization allows for faster processing of the data. The speed-up as compared to Isomap is roughly quadratic in the ratio between the number of quan- tizers and the number of data points. The quantizers and neighborhood structure are use to map the data to the low dimensional space.
Document type :
Conference papers
Complete list of metadatas

Cited literature [6 references]  Display  Hide  Download


https://hal.inria.fr/inria-00321500
Contributor : Jakob Verbeek <>
Submitted on : Wednesday, February 16, 2011 - 5:12:44 PM
Last modification on : Monday, September 25, 2017 - 10:08:04 AM
Long-term archiving on : Tuesday, May 17, 2011 - 2:32:40 AM

Files

verbeek02esann.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00321500, version 1

Citation

Jakob Verbeek, Nikos Vlassis, Ben Krose. Fast nonlinear dimensionality reduction with topology preserving networks. 10th Eurorean Symposium on Artificial Neural Networks (ESANN '02), Apr 2002, Bruges, Belgium. ⟨inria-00321500⟩

Share

Metrics

Record views

198

Files downloads

370