News

Abstract: Accurate uncertainty quantification for model-based, large-scale inverse problems represents one of the fundamental challenges in the context of computational science and engineering. In this thesis, novel Bayesian methodologies for the quantification of parametric and model…

Very excited to participate in this highly multidisciplinary project that aims at developing new methods for estimating local tumor infiltration and optimizing personalized treatment in glioma patients. More information can be found here.

...and the possibility of moving to a a blind-peer-review-based system

Our group will participate with three papers in this year's SIAM CSE 2017, namely: - Bayesian, Multi-Fidelity, Optimization under Uncertainty [slides] - Bayesian Coarse-Graining in Atomistic Simulations: Adaptive Identification of the Dimensionality and Salient Features [slides] - Probabilistic,…

Abstract: This presentation discusses the problem of optimization under uncertainty in a high-dimensional, numerically expensive setting and it's solution with limited computational resources. To this end we show how the problem of stochastic optimization can elegantly be rephrased as one of…

Please visit: http://www.tum-ias.de/bigdata2017/registration.html

Abstract: The present thesis is concerned with the topic of stochastic variational inference and its application to model-based Bayesian inverse problems. Variational inference, like Markov chain Monte Carlo (MCMC), is a method to evaluate intractable, complex prob- ability distributions. In…

Dec. 15-19 2016, University of Lugano, Switzerland

When: Wed. 29.06.2015, 10:00 Where: MW1403

Prof. Koutsourelakis has been invited to participate as a fellow in the group on Multiscale Modeling on Tumor Growth and Progression: From Gene regulation to Evolutionary Dynamics that will take place between Sept. 1 and Dec. 31 2016. More details on the group can be found at: …

Isabell Franck successfully defends her Ph.D. dissertation on "Sparse Variational Bayesian algorithms for large-scale inverse problems with applications in biomechanics".

Kickoff meeting of Radiotumor project at IGSSE - TUM

Some quick thoughts on SIAM CSE 2017

SIAM Computational Science and Engineering Conference, Atlanta Georgia 27.02-3.3.17

Maximilian Koschade presents his M.Sc. thesis on "Optimization under Uncertainty - A Bayesian, Multi-Fidelity Approach"

Registration is now open for "Machine Learning Challenges in Complex Multiscale Physical Systems"

Lukas Bruder presents his Semester thesis on "Stochastic variational Bayesian inference for inverse problems"

FOMICS Winter School on Uncertainty Quantification

Maximilian Koschade talks on "Information Geometry - Introduction to Theory and Application"

Prof. Koutsourelakis is invited as senior fellow at Center for Interdisciplinary Research (ZiF), Bielefeld University