Analysis of problems with incompletely known data by analytical and machine learning methods (Uncertain)
Main funder
Funder's project number: 349348
Funds granted by main funder (€)
- 33 800,00
Funding program
Project timetable
Project start date: 01/03/2022
Project end date: 29/02/2024
Summary
The aim of this project is to develop methods for quantifying the uncertainty of mathematical models. Here, we use three models that are typically used in real-life applications. These models are
(a) diffusion-type models combined with linear and non-linear reactions; and
(b) inverse models in X-ray tomography related to transport type equations.
In this study, we investigate how machine learning methods can be used to quantify uncertainty. We use analytical methods to validate results and train ML models. The aim is to show that analytical methods can be replaced by estimates produced by ML models.
(a) diffusion-type models combined with linear and non-linear reactions; and
(b) inverse models in X-ray tomography related to transport type equations.
In this study, we investigate how machine learning methods can be used to quantify uncertainty. We use analytical methods to validate results and train ML models. The aim is to show that analytical methods can be replaced by estimates produced by ML models.
