BCAM-IMUVA Summer School on Uncertainty Quantification for Applied Problems

July 4-7, 2016, Bilbao (Basque Country, Spain)

Home

Guggenheim Museum

Uncertainty Quantification (UQ) is a broad term that encompasses different methodologies, including uncertainty propagation, parameter estimation, model calibration, and error estimation.  The common goal of these methodologies is to assess how the uncertainties, both epistemic (lack of knowledge) and random (intrinsic variability), ubiquitous in all models, affect predictions and understanding of complex phenomena. Examples of intrinsic variability include uncertainty coming from inaccurate physical measurements, a bias in mathematical descriptions, as well as errors due to numerical approximations of computational simulations. The ability to predict the stochastic variations of a quantity of interest under variable conditions makes UQ essential for dealing with realistic experimental data and for the reliability of predictions based on numerical simulations.

The school aims at providing a survey of UQ and data assimilation techniques for some practical problems. It will be structured around four courses delivered by leading researchers in the field. Additionally, some participants will be given the opportunity of presenting their own results.

The school is addressed to mathematicians, statisticians, and scientists interested in UQ. PhD students and postdoctoral researchers attending the school may be financially supported by the organization.

The school will take place at Universidad de Deusto, Edificio "Comercial" ("La Comercial"), Ground floor, right, classroom no. 02, Bilbao (Basque Country, Spain), on July 4-7, 2016.

The school is organized by Elena Akhmatskaya, Mari Paz Calvo, Luca Gerardo-Giorda, and Jesús María Sanz-Serna.

Courses

  • Scalable Bayesian Inference with Hamiltonian Monte Carlo and Stan, by Dr. Michael Betancourt, University of Warwick (United Kingdom)
  • Algorithms for UQ for differential equations, by Prof. Max Gunzburger, Florida State University (USA)
  • Variational Assimilation and Uncertainty Quantification, by Prof. Olivier Talagrand, Laboratoire de Météorologie Dynamique, École Normale Supérieure (France) and Mohamed Jardak, Meteorological Office (United Kingdom)
  • Probabilistic and ensemble approaches to Data Assimilation, by Prof. Peter Jan van Leeuwen, University of Reading (United Kingdom)

Stan Tutorial

As a satellite event, Dr. Michael Betancourt has agreed to give a Tutorial on the Stan software package on July 8th, the day after the school will be closed. More details can be found here.


Organized by         Universidad de Valladolid     Universidad de Valladolid     Universidad de Valladolid     IMUVA