2025 CMS Summer Meeting
Quebec City, June 6 - 9, 2025
Je propose d’aborder cette réflexion pédagogique en m’appuyant sur des aspects de l’activité mathématique que je considère importants pour l’enseignement. Ces aspects ne sont pas nécessairement directement liés au contenu et objectifs spécifiques de chacun des cours, bien que leur importance relative puisse en dépendre fortement. J’aborderai des exemples tirés des cours que j’ai donnés et des activités de vulgarisation de l’AQJM (Association québécoise des jeux mathématiques), incluant des travaux d’équipe, des travaux de longue haleine et le « ungrading ».
Over 30 years, I have spent thousands of hours teaching, and thousands of hours thinking about teaching. Despite all this time and effort, every year I wonder about the best ways to maximize the positive impact for students engaged in any course that I teach. My reflection is ongoing, and with it a teaching which is alive, evolving. I can summarize my general goal by saying that I try to work so that students are fully invested in the doing of mathematics. What does that mean?
I propose to approach this pedagogical reflection by looking at aspects of mathematical activity that I consider important in relation to teaching. These aspects are not necessarily directly linked to the specific content and objectives of any given course, although their relative importance may well depend on it. I will look at concrete examples from courses I have taught and from outreach activities done by AQJM (Association québécoise des jeux mathématiques), including teamwork, long-term projects and ungrading.
The discretization of differential operators with respect to a small parameter h provides a discrete equation for the solution. Among other things we ask for convergence of the discrete solution to the actual solution as h-->0. However, in practice, h is finite. It turns out that preserving certain structures in the discretization provides great benefits for the approximated solution.
This talk will have two parts:
-1- I will give a short overview of what we mean by structure-preservation in the context of discretization of differential equations.
-2- I will look at the case of evolution PDEs which solutions develop a wide range of scales. I will show that by preserving the operator semi-group structure we may design a computationally efficient method with exceptional resolution properties. I will illustrate this technique with problems from fluid dynamics.