1 August 2006 in Applied Mathematics and Computation
1 October 2007 in Chaos, Solitons & Fractals
1 April 2006 in Applied Mathematics and Computation
1 January 2017 in SIAM Journal on Scientific Computing
We present an algorithm for numerical computations involving trivariate functions in a three-dimensional rectangular parallelepiped in the context of Chebfun. Our scheme is based on low-rank representation through multivariate adaptive cross approximation. The component one-dimensional functions are represented by finite Chebyshev expansions, or trigonometric expansions in the periodic case. Numerical experiments show the power and convenience of Chebfun3 for problems such as function manipulation, differentiation, optimization, and integration, as well as for exploration of fundamental issues of multivariate approximation and low-rank compression.
1 November 2006 in Applied Mathematics and Computation
1 January 2010 in SIAM Journal on Matrix Analysis and Applications
We present methods to compute verified square roots of a square matrix A. Given an approximation X to the square root, obtained by a classical floating point algorithm, we use interval arithmetic to find an interval matrix which is guaranteed to contain the error of X. Our approach is based on the Krawczyk method, which we modify in two different ways in such a manner that the computational complexity for an $n\times n$ matrix is reduced to $n^3$. The methods are based on the spectral decomposition or, in the case that the eigenvector matrix is ill conditioned, on a similarity transformation to block diagonal form. Numerical experiments prove that our methods are computationally efficient and that they yield narrow enclosures provided X is a good approximation. This is particularly true for symmetric matrices, since their eigenvector matrix is perfectly conditioned.