Grant: Discovery Grants Program - Individual
Dalhousie University
$31,500.00
Nonlinear and computationally analysis
EF 2008-09
Résumé
Nonlinear and Computationally Assisted Analysis. My intention is to further develop my long-term program in nonlinear functional analysis and optimization in conjunction with my Canada Research Chair and CFI funded research on computer-assisted discovery. Over the past four years at Dalhousie I have been able to start a laboratory for Distributed Research and Collaborative Technology (Dalhousie Distributed Research Institute and Virtual Environment, http://projects.cs.dal.ca/ddrive/). The Ddrive includes a newly installed 240 core mathematical computing cluster whose primary aim is to provide advanced (parallel) mathematical computing resources in my lab and across Canada. Unlike conventional 'capacity' clusters, the emphasis here is on rapid insight generation provided by easy to use 'hybrid' (numerical/symbolic) parallel software. This work is in part sponsored by MapleSoft. The fruits of this programme so far are described in three recent books on Experimental Mathematics with David Bailey (Lawrence Berkeley) and various other co-authors.
My contention is that this approach is applicable across mathematics and my aim over the next five years is to further substantiate the value of such inductive methods. For many applications in the sciences we have found that a prerequisite is to be able to reliably compute multidimensional integrals to very high precision (over 100 digits). Hybrid (numeric-symbolic) parallel techniques based on doubly-exponential quadrature methods are now allowing us, often with Richard Crandall (Apple senior scientist) frequently to do this in as many as five or six dimensions; but much algorithmic refinement is still needed, especially in the matter of certifying at least some digits of the answers (for which Bailey and I have some non trivial partial results). At the same time, I continue my long-term studies in convex and nonlinear functional analysis, with emphasis on the structure of maximal monotone operators in non-reflexive space.
Programme
Discovery Grants Program - Individual
Bénéficiaire
Emplacement
Nova Scotia, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
139620-2008
Name nom
Borwein, Jonathan
Department département
Computer Science, Faculty of
Organization id
79
Institution établissement
Dalhousie University
Province en
Nova Scotia
Province fr
Nouvelle-Écosse
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2008
Competition year année de concours
2008
Award amount
31500
Program id
RGPIN
Program nane en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
Discovery grants programs
Group fr
Programmes de subventions à la découverte
Committee code
336
Committee name en
Pure & Applied Maths - A
Committee name fr
Maths pures et appliquées - A
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
2901
Research subject group en
Pure mathematics
Research subject group fr
Mathématiques pures
Research subject en
Classical analysis
Research subject fr
Analyse classique
Partie
1
Nb partie
5
Application title
Nonlinear and computationally analysis
Keyword
NA
Application summary
Nonlinear and Computationally Assisted Analysis. My intention is to further develop my long-term program in nonlinear functional analysis and optimization in conjunction with my Canada Research Chair and CFI funded research on computer-assisted discovery. Over the past four years at Dalhousie I have been able to start a laboratory for Distributed Research and Collaborative Technology (Dalhousie Distributed Research Institute and Virtual Environment, http://projects.cs.dal.ca/ddrive/). The Ddrive includes a newly installed 240 core mathematical computing cluster whose primary aim is to provide advanced (parallel) mathematical computing resources in my lab and across Canada. Unlike conventional 'capacity' clusters, the emphasis here is on rapid insight generation provided by easy to use 'hybrid' (numerical/symbolic) parallel software. This work is in part sponsored by MapleSoft. The fruits of this programme so far are described in three recent books on Experimental Mathematics with David Bailey (Lawrence Berkeley) and various other co-authors.
My contention is that this approach is applicable across mathematics and my aim over the next five years is to further substantiate the value of such inductive methods. For many applications in the sciences we have found that a prerequisite is to be able to reliably compute multidimensional integrals to very high precision (over 100 digits). Hybrid (numeric-symbolic) parallel techniques based on doubly-exponential quadrature methods are now allowing us, often with Richard Crandall (Apple senior scientist) frequently to do this in as many as five or six dimensions; but much algorithmic refinement is still needed, especially in the matter of certifying at least some digits of the answers (for which Bailey and I have some non trivial partial results). At the same time, I continue my long-term studies in convex and nonlinear functional analysis, with emphasis on the structure of maximal monotone operators in non-reflexive space.