BannerHauptseite TUMHauptseite LehrstuhlMathematik SchriftzugHauptseite LehrstuhlHauptseite Fakultät

Reading Group Uncertainty Quantification

The reading group meets in the Maths/Computer Science building on the Garching campus. Everybody is welcome to attend. Subscription to the group's mailing list is possible here.

Important information: Due to the ongoing coronavirus pandemic we meet and discuss ONLINE using Zoom. The moderator will create a meeting link and password, and distribute it via the group mailing list.

Sommersemester 2022

Date Time Moderator Topic
05.07.22 16:00 Laura Scarabosio (Radboud) Data driven gradient flows
21.06.22 16:00 Daniel Walter (Linz) Sparse solutions in optimal control of PDEs with uncertain parameters: the linear case
07.06.22 16:00 Abdul-Lateef Haji-Ali (Heriot-Watt, Edinburgh) Unbiased Multilevel Monte Carlo methods for intractable distributions: MLMC meets MCMC
24.05.22 16:00 Giovanni Rabitti (Heriot-Watt, Edinburgh) On Shapley Value for Measuring Importance of Dependent Inputs
10.05.22 16:00 Florian Beiser (NTNU) Combining data assimilation and machine learning to infer unresolved scale parametrization

Wintersemester 2021/22

Date Time Moderator Topic
08.02.22 16:00 Felipe Uribe (DTU, Copenhagen) Unbiased Markov chain Monte Carlo with couplings
25.01.22 16:00 Björn Sprungk (Freiberg) Analysis of a class of Multi-Level Markov Chain Monte Carlo algorithms based on Independent Metropolis-Hastings
11.01.22 16:00 Jonas Latz (Heriot-Watt, Edinburgh) Probability, Frequency and Reasonable Expectation
07.12.21 16:00 Leila Taghizadeh (TUM) Optimal experimental design under irreducible uncertainty for linear inverse problems governed by PDEs
23.11.21 16:00 Fabian Wagner (TUM) Physics-Informed Machine Learning with Conditional Karhunen-Loève Expansions
09.11.21 16:00 Elisabeth Ullmann (TUM) Unbiased MLMC-based variational Bayes for likelihood-free inference
26.10.21 16:00 Simon Urbainczyk (Heriot-Watt, Edinburgh) Analysis of boundary effects on PDE-based sampling of Whittle-Matérn random fields
12.10.21 16:00 Jan Stanczuk (Cambridge) Score-Based Generative Modeling through Stochastic Differential Equations

Sommersemester 2021

Date Time Moderator Topic
20.07.21 16:00 Simon Weißmann (Heidelberg) Consensus Based Sampling
06.07.21 16:00 Laura Scarabosio (Radboud) Estimates on the generalization error of physics-informed neural networks for approximating a class of inverse problems for PDEs
22.06.21 16:00 Fabian Wagner (TUM) Complete dynamics and spectral decomposition of the Ensemble Kalman Inversion
08.06.21 16:00 Jonas Latz (Cambridge) A dynamical systems framework for intermittent data assimilation
25.05.21 16:00 Daniel Schaden (TUM) On quantitative stability in infinite-dimensional optimization under uncertainty
11.05.21 16:00 Sam Power (Bristol) Transport map accelerated Markov chain Monte Carlo
27.04.21 16:00 Philipp Eisenhauer (Bonn) Structural models for policy-making: Coping with parametric uncertainty

Wintersemester 2020/21

Date Time Moderator Topic
09.02.21 16:00 Felipe Uribe (DTU, Copenhagen) A Stochastic Newton MCMC Method for Large-Scale Statistical Inverse Problems with Application to Seismic Inversion
15.12.20 16:00 Jonas Latz (Cambridge) The linear conditional expectation in Hilbert space
01.12.20 16:00 Melina Freitag (Potsdam) Uncertainty quantification in large Bayesian linear inverse problems using Krylov subspace methods
17.11.20 16:00 Elisabeth Ullmann (TUM) Robust Training and Initialization of Deep Neural Networks: An Adaptive Basis Viewpoint
03.11.20 16:00 Fabian Wagner (TUM) Interacting Langevin Diffusions: Gradient Structure And Ensemble Kalman Sampler
20.10.20 16:00 Laura Scarabosio (Radboud) Bayesian mesh adaptation for distributed parameters

Sommersemester 2020

Date Time Moderator Topic
07.07.20 16:00 Simon Weißmann (Mannheim) Tikhonov Regularization Within Ensemble Kalman Inversion
23.06.20 16:00 Florian Beiser (TUM) Risk-Averse PDE-Constrained Optimization Using the Conditional Value-At-Risk
09.06.20 16:00 Robert Scheichl (Heidelberg) On the geometry of Stein variational gradient descent
26.05.20 16:00 Daniel Schaden (TUM) Continuous Level Monte Carlo and Sample-Adaptive Model Hierarchies
12.05.20 16:00 Felipe Uribe (DTU, Copenhagen) Cauchy difference priors for edge-preserving Bayesian inversion
28.04.20 16:00 Jonas Latz (Cambridge) Calibrate, Emulate, Sample

Wintersemester 2019/20

Date Time Room Moderator Topic
28.01.20 16:00 02.10.011 Fabian Wagner Stein Variational Gradient Descent: A General Purpose Bayesian Inference Algorithm
14.01.20 16:45 02.10.011 Simon Urbainczyk Adaptive Sampling Strategies for Stochastic Optimization
17.12.19 16:00 02.10.011 Ustim Khristenko Bayesian Deep Convolutional Encoder-Decoder Networks for Surrogate Modeling and Uncertainty Quantification
19.11.19 16:00 02.10.011 Mario Parente The Lipschitz matrix: a tool for parameter space dimension reduction
05.11.19 16:00 02.10.011 Fabian Wagner Estimation of small failure probabilities in high dimensions by subset simulation
22.10.19 16:00 02.10.011 Jonas Latz Why Are Big Data Matrices Approximately Low Rank?

Sommersemester 2019

Date Time Room Moderator Topic
30.07.19 16:00 02.10.011 Yannik Schälte (ICB) ABC Samplers
16.07.19 16:00 02.10.011 Brendan Keith (M2) Optimization of PDEs with uncertain inputs
25.06.19 16:00 02.10.011 Florian Beiser (M2) A stochastic gradient method with mesh refinement for PDE constrained optimization under uncertainty
18.06.19 16:00 02.10.011 Mario Parente (M2) A transport-based multifidelity preconditioner for Markov chain Monte Carlo
28.05.19 16:00 02.10.011 Laura Scarabosio (M2) A Multiscale Strategy for Bayesian Inference Using Transport Maps
14.05.19 16:15 02.10.011 Elisabeth Ullmann (M2) Conditional Karhunen-Loeve expansion for uncertainty quantification and active learning in partial differential equation models
30.04.19 16:15 02.10.011 Ionut-Gabriel Farcas (I5) Sensitivity-driven adaptive sparse stochastic approximations in plasma microinstability analysis

Wintersemester 2018/19

Date Time Room Moderator Topic
19.02.19 16:15 02.10.011 Konstantin Riedl (TUM) Multifidelity Preconditioning of the Cross-Entropy Method for Rare Event Simulation and Failure Probability Estimation
05.02.19 16:15 02.10.011 Jonas Latz (M2) How Deep are Deep Gaussian processes?
22.01.19 16:15 02.10.011 Fabian Wagner (M2) Physics-Informed Generative Adversarial Networks for Stochastic Differential Equations
08.01.19 16:15 02.10.011 Steven Mattis (M2) A Fast and Scalable Method for A-Optimal Design of Experiments for Infinite-dimensional Bayesian Nonlinear Inverse Problems
20.11.18 16:15 02.10.011 Ustim Khristenko (M2) Analysis of boundary effects on PDE-based sampling of Whittle-Matérn random fields
06.11.18 16:15 02.10.011 Laura Scarabosio (M2) Well-posed Bayesian geometric inverse problems arising in subsurface flow
23.10.18 16:15 02.10.011 Mario Parente (M2) A probabilistic framework for approximating functions in the active subspace

Sommersemester 2018

Date Time Room Moderator Topic
24.07.18 16:15 02.10.011 Sabrina Balzer (M2) Posterior Consistency for Gaussian Process Approximations of Bayesian Posterior Distributions
10.07.18 16:15 02.10.011 Mario Parente (M2) Importance Sampling and Necessary Sample Size: An Information Theory Approach
26.06.18 16:15 02.10.011 Daniel Schaden (M2) MCMC Methods for Functions: Modifying Old Algorithms to Make Them Faster
12.06.18 16:15 02.10.011 Ludger Pähler (TUM) PDE-Net: Learning PDEs from Data
29.05.18 16:15 02.10.011 Elisabeth Ullmann (M2) Efficient white noise sampling and coupling for multilevel Monte Carlo with non-nested meshes
15.05.18 16:15 02.10.011 Jonas Latz (M2) Deep Learning: An Introduction for Applied Mathematicians
24.04.18 16:15 02.10.011 Steven Mattis (M2) Combining Push-Forward Measures and Bayes' Rule to Construct Consistent Solutions to Stochastic Inverse Problems

Wintersemester 2017/18

Date Time Room Moderator Topic
30.01.18 16:15 02.10.011 Elisabeth Ullmann (M2) Machine learning of differential equations using Gaussian processes (Paper 1) (Paper 2)
16.01.18 16:15 02.10.011 Jonas Latz (M2) Statistical analysis of differential equations: introducing probability measures on numerical solutions
05.12.17 16:15 02.10.011 Steven Mattis (M2) A Short Course on Duality, Adjoint Operators, Green’s Functions, and A Posteriori Error Analysis (§3-4)
21.11.17 16:15 02.10.011 Steven Mattis (M2) A Short Course on Duality, Adjoint Operators, Green’s Functions, and A Posteriori Error Analysis (§1-2)
07.11.17 16:15 02.10.011 Laura Scarabosio (M2) Finite element methods for semilinear elliptic stochastic partial differential equations
24.10.17 16:15 02.10.011 Piotr Swierczynski (M2) An explicit link between Gaussian fields and Gaussian Markov random fields: the stochastic partial differential equation approach

Sommersemester 2017

Date Time Room Moderator Topic Slides
12.04.17 14:00 02.08.011 Laura Scarabosio (M2) A Fresh Look at the Kalman Filter  
02.05.17 16:15 03.06.011 Mario Parente (M2) Ensemble Kalman methods for inverse problems pdf, proofs
16.05.17 16:15 03.06.011 Ionut Farcas (CS) An efficient Bayesian inference approach to inverse problems based on an adaptive sparse grid collocation method  
06.06.17 16:15 03.06.011 Jonas Latz (M2) Analysis of the ensemble Kalman filter for inverse problems  
27.06.17 16:15 03.06.011 Elisabeth Ullmann (M2) On the Brittleness of Bayesian Inference  
11.07.17 16:15 03.06.011 Mario Parente (M2) Active Subspaces: Emerging Ideas for Dimension Reduction in Parameter Studies (§1-3) pdf
25.07.17 16:15 03.06.011 Steven Mattis (M2) Active Subspaces: Emerging Ideas for Dimension Reduction in Parameter Studies (§4-6)  
23.08.17 10:00 03.06.011 Daniel Schaden (M2) Optimal Model Management for Multifidelity Monte Carlo Estimation  

Wintersemester 2016/17

Date Time Room Moderator Topic Slides Further Material
17.11.16 10:00 03.11.018 Jonas Latz (M2) Stuart - Uncertainty Quantification in Bayesian Inversion Bayesian Statistics
08.12.16 10:15 03.11.018 Elisabeth Ullmann (M2) Stuart(2010) - Inverse Problems: A Bayesian Perspective (§6)  
15.12.16 10:15 03.11.018 Elisabeth Ullmann (M2) Stuart(2010) - Inverse Problems: A Bayesian Perspective (§6) Random fields
19.01.17 10:15 03.11.018 Tinsley Oden (ICES) OPAL: the Occam Plausibility Algorithm for Bayesian Model Selection and Validation  
02.02.17 10:15 03.11.018 Steven Mattis (M2) Stuart(2010) - Inverse Problems: A Bayesian Perspective (§1-3) Bayesian Inverse
22.02.17 14:15 - Jonas Latz (M2) Stuart(2010) - Inverse Problems: A Bayesian Perspective (§4-5) Well-posed problems, Algorithms GitHub Implementations