Kernel methods & RBFs
Stable bases, greedy point selection, rational and variably scaled kernels, meshless PDE methods.
Read moreUniversity of Padova · Department of Medicine
Stefano
De Marchi
Full Professor of Numerical Analysis
I work at the intersection of approximation theory, radial basis functions, scientific computing and data-driven applications in imaging and medicine.
Numerical analysis,
approximation, geometry and computation
approximation, geometry and computation
Current appointmentDepartment of Medicine, University of Padova
International affiliationDepartment of Mathematics, Selçuk University
Scientific networksUMI–TAA · GNCS–INdAM · PNC
Research
Mathematics that moves between theory and application.
Polynomial approximation
Padua points, admissible meshes, Fekete and Leja points, multivariate interpolation and cubature.
Read moreDiscontinuities & imaging
Mapped bases, fake nodes and discontinuous kernels for Gibbs mitigation and image reconstruction.
Read moreData, topology & learning
Persistence kernels, intrinsic dimension, neural approximation and computational medicine.
Read moreSelected recent work
New directions in approximation and computation.
Learning Radial Basis Functions from the German School: A Personal Perspective
Engineering Analysis with Boundary Elements, Vol. 191, 106924.
Approximation with rational and rescaled kernels: new steps forward
With Elisabeth Larsson · Modern Mathematical Methods.
Polynomial approximation from diffused data: unisolvence and stability
With Ludovico Bruni Bruno and Giacomo Elefante · BIT Numerical Mathematics.
Academic community
Research grows through networks.
Coordination, editorial activity and international collaborations are a central part of my academic work.
“
On the roads of justice and love, the Christian who does not give more is lost.Don Primo Mazzolari