Where and when
April 5, Tuesday · 12:00 – 1:00PM @ Via di Vallerano 139, Sala Consiglio or via Google Meet: https://meet.google.com/eje-dpyz-apu
Abstract
In this talk we provide an overview of the methods that can be used for prediction under uncertainty and parameter calibration of dynamical systems, and of the fundamental challenges that arise in this context. In particular, we raise a warning flag about identifiability of the parameters of ODE-based models; often, it might be hard to infer the correct values of the parameters from data, even for very simple models, making it non-trivial to use these models for meaningful predictions. Most of the points that we touch upon are actually generally valid for inverse problems in general setups, and can be adapted to the case of PDE-based models.
Short bio
Chiara Piazzola got a M.Sc. in Mathematics at the University of Verona, and a Ph.D. in Mathematics at University of Innsbruck on low-rank approximations of high-dimensional problems. She’s been a post-doc at CNR-IMATI since March 2020, where she works on numerical methods for uncertainty quantification.