Workshop
Bayesian methods for uncertainty quantification
Wednesday 23 October 10.30
Organizer: Rémi Laumont, Technical University of Denmark
Bayesian methods for uncertainty quantification (UQ) are increasingly recognized as a powerful approach to managing uncertainty. At their core, Bayesian methods are about updating our beliefs in light of new evidence. At first we start with a certain level of belief about a situation. As new information becomes available, Bayesian methods refine and adjust this belief. These methods have several advantages. They firstly incorporate prior knowledge or expert opinions.
They also describe the full range of possible outcomes with their associated probabilities. It is crucial for informed decision-making. In a world where data is abundant but often incomplete or uncertain, the capability to integrate prior knowledge and update beliefs with new data is valuable. This topic is particularly relevant in data science where Bayesian UQ stands out as a critical approach for making informed, data-driven decisions. Finally, it has a various and diverse range of application in geoscience or medicine eg.
This mini-symposium focuses both on one hand on the recent advances in AI based Bayesian science and on the other hand on practice with real-world examples.
In this mini-symposium, we target a broad audience from the Bayesian theorist to the non-expert scientist.