Inverse problems
Mathematical imaging, blind reconstruction and uncertainty quantification.
Research at the interface of mathematics, modeling, imaging and biology
I develop mathematical models and scalable algorithms for computational imaging, with applications in microscopy, MRI and satellite imaging. My work connects optimization, approximation theory, Bayesian inference and neural networks.
Research
Core themes
Optimization
Fast algorithms for large-scale, non-convex and measure-space problems.
Learning and inference
Neural networks, learned priors, Bayesian inference and operator approximation.
Scientific imaging
Models and software for microscopy, MRI, satellite imaging and biology.
Current work
Recent highlights
Convex Curve Recovery in Phase Space
Extreme points and exact graph discretization.
Read the paperAn analytic theory of convolutional neural network inverse problem solvers
Read the paperJackpot: Approximating Uncertainty Domains with Adversarial Manifolds
With N. Munier and E. Soubies.
Read the paperDeepInverse
A Python package for solving imaging inverse problems with deep learning.
View documentationContact
Let’s discuss research.
pierre.weiss@cnrs.fr
Centre de Biologie Intégrative
Bâtiment 4R4, bureau 2034
169 rue Marianne Grunberg-Manago
31400 Toulouse, France