Sciweavers

Share
ISMIR
2005
Springer

Exploiting the Tradeoff Between Precision and Cpu-Time to Speed Up Nearest Neighbor Search

9 years 4 months ago
Exploiting the Tradeoff Between Precision and Cpu-Time to Speed Up Nearest Neighbor Search
We describe a recursive algorithm to quickly compute the N nearest neighbors according to a similarity measure in a metric space. The algorithm exploits an intrinsic property of a large class of similarity measures for which some parameter p has a positive influence both on the precision and the cpu cost (precision-cputime tradeoff). The algorithm uses successive approximations of the measure to compute first cheap distances on the whole set of possible items, then more and more expensive measures on smaller and smaller sets. We illustrate the algorithm on the case of a timbre similarity algorithm, which compares gaussian mixture models using a Monte Carlo approximation of the Kullback-Leibler distance, where p is the number of points drawn from the distributions. We describe several Monte Carlo algorithmic variants, which improve the convergence speed of the approximation. On this problem, the algorithm performs more than 30 times faster than the naive approach.
Pierre Roy, Jean-Julien Aucouturier, Franço
Added 27 Jun 2010
Updated 27 Jun 2010
Type Conference
Year 2005
Where ISMIR
Authors Pierre Roy, Jean-Julien Aucouturier, François Pachet, Anthony Beurivé
Comments (0)
books