Finding the nearest neighbor among a large collection of high dimensional vectors can be a computationally demanding task. In this paper, we pursue fast vector matching by representing vectors in with lower dimensional projections in
,
. The key to creating and using the representative vectors is a lower bound on the Euclidean distance between arbitrary vectors in
based on the submultiplicative property of induced matrix norms. For any non-zero projection matrix
, the bound is proportional to the distance between the projected vectors. We study other existing bounds involving orthogonal transforms and piecewise constant approximation maps in light of this formulation. Additionally, we address the question of how to optimize the projection matrix given a dataset in order to make the bound as tight as possible. Experimental results on a speech database show that exact nearest neighbor computation can be accelerated by a factor of 5 using the proposed bound.
By: George Saon; Peder Olsen
Published in: RC24859 in 2009
LIMITED DISTRIBUTION NOTICE:
This Research Report is available. This report has been submitted for publication outside of IBM and will probably be copyrighted if accepted for publication. It has been issued as a Research Report for early dissemination of its contents. In view of the transfer of copyright to the outside publisher, its distribution outside of IBM prior to publication should be limited to peer communications and specific requests. After outside publication, requests should be filled only by reprints or legally obtained copies of the article (e.g., payment of royalties). I have read and understand this notice and am a member of the scientific community outside or inside of IBM seeking a single copy only.
Questions about this service can be mailed to reports@us.ibm.com .