Home | Bookmarks | Papers | Blog |

In this paper, we show that there exists a small core-set for the
problem of computing the ``smallest'' radius **k**-flat for a given
point-set in **R ^{d}**. The size of the core-set is

No previous efficient approximation algorithms were known
for those problems in high-dimensions, when **k>1** or **j>1**.

Abstract - Postscript, PDF.

source (latex).

@string{SOCG_2002 = "Proc. 18th Annu. ACM Sympos. Comput. Geom."} @inproceedings{hv-pchdu-02, author = "S. {Har-Peled} and K. R. Varadarajan", booktitle = SOCG_2002, title = "Projective Clustering in High Dimensions using Core-Sets", year = 2002, pages = {312--318} }

Last modified: Fri Jul 7 18:10:40 EDT 2000