The Fast Gauss Transform
Abstract
Many problems in applied mathematics require the evaluation of the sum of N Gaussians at M points in space. The work required for direct evaluation grows like N M as N and M increase; this makes it very expensive to carry out such calculations on a large scale. In this paper, we present an algorithm which evaluates the sum of N Gaussians at M arbitrarily distributed points in C (dot) (N + M) work, where C depends only on the precision required. When N = M = 100, 000, our algorithm is several thousand times faster than direct evaluation. It is based on a divide and conquer strategy, combined with the manipulation of Hermite expansions and Taylor series. Keywords: Gauss transform; Hermite polynomials; Fast algorithms.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 01, 1989
- Accession Number
- ADA211287
Entities
People
- J. Strain
- L. Greengard
Organizations
- Yale University