Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

 
Boston University Computer Science Technical Report No. 2004014, April 12, 2004. Learning Euclidean Embeddings for Indexing and
 

Summary: Boston University Computer Science Technical Report No. 2004­014, April 12, 2004.
Learning Euclidean Embeddings for Indexing and
Classification
Vassilis Athitsos, Joni Alon, Stan Sclaroff, and George Kollios #
Computer Science Department
Boston University
111 Cummington Street
Boston, MA 02215, USA
email: {athitsos, jalon, sclaroff, gkollios}@cs.bu.edu
ABSTRACT
BoostMap is a recently proposed method for e#cient ap­
proximate nearest neighbor retrieval in arbitrary non­ Eu­
clidean spaces with computationally expensive and possibly
non­metric distance measures. Database and query objects
are embedded into a Euclidean space, in which similarities
can be rapidly measured using a weighted Manhattan dis­
tance. The key idea is formulating embedding construc­
tion as a machine learning task, where AdaBoost is used
to combine simple, 1D embeddings into a multidimensional
embedding that preserves a large amount of the proximity

  

Source: Athitsos, Vassilis - Department of Computer Science and Engineering, University of Texas at Arlington
Boston University, Department of Computer Science

 

Collections: Computer Technologies and Information Sciences