Canada Spends Logo

Explore our Budget 2026 analysis

Grant: Discovery Grants Program - Individual

University of Manitoba

$28,000.00

Fast orthogonal polynomial transforms

FY 2024-25

Summary
Orthogonal polynomial expansions have important applications that require the transforms of synthesis and analysis to convert between coefficient- and grid-space representations. Classical examples include spherical harmonic transforms, currently used for numerical weather prediction, Zernike polynomials on the disk in optics and fluid dynamics, and Proriol polynomials on the triangle that play a central role in the spectral element method where partial differential equations are solved in infinite dimensions on each triangle in a mesh. My widely used free and open-source software packages in Julia and in C contain implementations of my work in this area and will see contributions from the proposed research described below. I have developed a collection of fast orthogonal polynomial transforms for classical orthogonal polynomials and multivariate orthogonal polynomials from the Koornwinder construction, including the three examples above. I presented a unified approach as a co-author in an Acta Numerica review that characterizes the matrices of connection coefficients between classical orthogonal polynomials as generalized eigenvector matrices of upper-triangular or symmetric-definite and banded generalized eigenvalue problems. In recent work, I describe fast algorithms for approximating the connection coefficients between a family of orthogonal polynomials and another family with a polynomially or rationally modified measure. The connection coefficients are computed via infinite-dimensional banded matrix factorizations and may be used to compute the modified Jacobi matrices all in linear complexity with respect to the truncation degree. This recent work has opened numerous avenues of future research, which I intend to pursue. I plan to explore fast orthogonal transforms beyond the classical univariate regime and the original conditions in the Koornwinder construction. The main research objective is to computationally explore orthogonal polynomials inside multivariate regions with curved boundaries that can be well-approximated by polynomials or algebraic curves. This will enable the construction of a sparse spectral element method on such regions with optimal complexities in building partial differential operators and solving linear systems to approximate solutions of partial differential equations. When fulfilled, this research programme will have an outsized impact on the utility of spectral methods for real-world applications. This programme will result in the training of two M Sc students, three Ph D students, two postdoctoral researchers and up to five undergraduate student researchers over the summers. The trainees will be well-equipped to work or continue their studies in the natural sciences and engineering disciplines.
Program
Discovery Grants Program - Individual
Location
Manitoba, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data

As published

Application id
RGPIN-2024-05480
Name nom
Slevinsky, Richard
Department département
Mathematics
Organization id
19
Institution établissement
University of Manitoba
Province en
Manitoba
Province fr
Manitoba
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2024
Award amount
28000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1508
Committee name en
Mathematics and Statistics
Committee name fr
Mathématiques et statistique
Area of application code
1205
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Mathematical sciences
Area of application fr
Sciences mathématiques
Research subject code
2955
Research subject group en
Applied mathematics
Research subject group fr
Mathématiques appliquées
Research subject en
Numerical analysis
Research subject fr
Analyse numérique
Field of research list names en
Not available
Field of research list names fr
Non disponible
Socioeconomic objective list names en
Not available
Socioeconomic objective list names fr
Non disponible
Application title
Fast orthogonal polynomial transforms
Keywords
orthogonal polynomials; fast transforms; sparse spectral methods; polynomial and rational approximation theory; special functions; shared memory parallelism; fast multipole method; fast fourier transform; divide-and-conquer eigenvalue problems
Application summary
Orthogonal polynomial expansions have important applications that require the transforms of synthesis and analysis to convert between coefficient- and grid-space representations. Classical examples include spherical harmonic transforms, currently used for numerical weather prediction, Zernike polynomials on the disk in optics and fluid dynamics, and Proriol polynomials on the triangle that play a central role in the spectral element method where partial differential equations are solved in infinite dimensions on each triangle in a mesh. My widely used free and open-source software packages in Julia and in C contain implementations of my work in this area and will see contributions from the proposed research described below. I have developed a collection of fast orthogonal polynomial transforms for classical orthogonal polynomials and multivariate orthogonal polynomials from the Koornwinder construction, including the three examples above. I presented a unified approach as a co-author in an Acta Numerica review that characterizes the matrices of connection coefficients between classical orthogonal polynomials as generalized eigenvector matrices of upper-triangular or symmetric-definite and banded generalized eigenvalue problems. In recent work, I describe fast algorithms for approximating the connection coefficients between a family of orthogonal polynomials and another family with a polynomially or rationally modified measure. The connection coefficients are computed via infinite-dimensional banded matrix factorizations and may be used to compute the modified Jacobi matrices all in linear complexity with respect to the truncation degree. This recent work has opened numerous avenues of future research, which I intend to pursue. I plan to explore fast orthogonal transforms beyond the classical univariate regime and the original conditions in the Koornwinder construction. The main research objective is to computationally explore orthogonal polynomials inside multivariate regions with curved boundaries that can be well-approximated by polynomials or algebraic curves. This will enable the construction of a sparse spectral element method on such regions with optimal complexities in building partial differential operators and solving linear systems to approximate solutions of partial differential equations. When fulfilled, this research programme will have an outsized impact on the utility of spectral methods for real-world applications. This programme will result in the training of two M Sc students, three Ph D students, two postdoctoral researchers and up to five undergraduate student researchers over the summers. The trainees will be well-equipped to work or continue their studies in the natural sciences and engineering disciplines.