Topic selection & color key :

Topics

All topics are shown.

Forms of presentation

Forms of presentation

All types are shown.

SIAM Conference on Uncertainty Quantification (UQ20)

Thursday – 26.03.2020

08:30

iCal
Thordis Thorarinsdottir | Norwegian Computing Center | Norway

MT05: Thordis Thorarinsdottir: Spatial statistics: Modelling spatial variation

Room:
MW HS 2001

Topic:

Form of presentation:
Mini-tutorial

Duration:
120 Minutes

08:30

Spatial statistics: Modelling spatial variation

08:30

iCal
John Jakeman | Sandia National Laboratories | United States

MT06: John Jakeman: Multi-fidelity UQ: How to use model ensembles of varying cost and accuracy

Room:
MW HS 0001

Topic:
Uncertainty propagation

Form of presentation:
Mini-tutorial

Duration:
120 Minutes

08:30

- CANCELED - Multi-fidelity UQ: How to use model ensembles of varying cost and accuracy

08:30

iCal
Jonathan Hobbs | Jet Propulsion Laboratory, California Institute of Technology | United States

Amy Braverman | Jet Propulson Laboratory, California Institute of Technology | United States

Joaquim Teixeira | Jet Propulsion Laboratory, California Institute of Technology | United States

Meredith Franklin | University of Southern California | United States

MS441: Uncertainty Quantification for Earth Remote Sensing (Part I of III)

Room:
MW HS 1801

Topic:
Inverse problems

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Numerous Earth-observing satellites provide high-resolution and high-volume data that facilitate scientific inference on physical and environmental processes. Most remote sensing data products used for scientific investigations are often subject to multiple stages of processing before they are widely used, and the scientific utility of these data products critically depends on a comprehensive assessment of the sources of uncertainty encountered in these stages of processing. One key stage involves the use of a retrieval algorithm to infer a geophysical quantity of interest from a satellite’s observed intensity of radiation.

The retrieval is an inverse problem that has been implemented mathematically and computationally in numerous ways for different satellite missions. Several of the presentations in this mini-symposium will each highlight an individual Earth-observing satellite and its retrieval methodology, emphasizing important contributions to uncertainty in retrieval data products. Methodological developments that interrogate the joint distribution of true geophysical states, retrieved states, and observed satellite spectra will be introduced. The presentations will span multiple Earth science applications, including weather and climate, the carbon cycle, air quality, atmospheric chemistry, and ecosystem health.

08:30

Uncertainty quantification for temperature and humidity retrievals from atmospheric sounders

09:00

Multivariate spatial statistical modeling and applications in simulation-based uncertainty experiments for remote sensing

09:30

Assessing the impact of uncertainty in the Orbiting Carbon Observatory-2 estimates of CO2 concentration

10:00

An uncertainty quantification framework for optimal estimation retrieval of aerosols

08:30

iCal
Vesa Kaarnioja | University of New South Wales | Australia

Yoshihito Kazashi | École polytechnique fédérale de Lausanne | Switzerland

Dirk Nuyens | KU Leuven | Belgium

MS033: Kernel, Quasi-Monte Carlo, and Sparse Grid Methods for High-dimensional Approximation and Integration (Part III of III)

Chair(s)
Alexander Gilbert (Universität Heidelberg)

Fabio Nobile (EPFL)

Michael Griebel (University of Bonn and Fraunhofer SCAI)

Yoshihito Kazashi (EPFL)

Fabio Nobile (EPFL)

Michael Griebel (University of Bonn and Fraunhofer SCAI)

Yoshihito Kazashi (EPFL)

Room:
MW HS 0350

Topic:
High-dimensional approximation

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Organized in co-operation with

GAMM AG UQ

This mini-symposium aims at bringing together people working on kernel and other sampling-based approximation methods for high-dimensional problems, in particular, but not restricted to, quasi-Monte Carlo methods, and sparse grids methods. Kernel methods and the related Gaussian Process surrogate models are a powerful class of numerical methods, and they are often employed in problems arising in uncertainty quantification. Nonetheless, there is much to be explored in their theoretical analysis for UQ applications, which are often formulated as high-dimensional approximation or integration problems.

On the other hand, the theory and applicability of QMC and sparse grid approximation/integration techniques in high or infinite dimensional problems have seen considerable advances in the last years, yet being far from addressing all problems of interest in UQ.

The objective of this mini-symposium is to showcase the late theoretical results and exchange ideas on sampling-based high dimensional integration and approximation methods targeting UQ applications.

08:30

Higher order quasi-Monte Carlo rules for uncertainty quantification using periodic random variables

09:00

Analysis of the dynamical low rank equations for random semi-linear parabolic problems

09:30

MDFEM for elliptic PDEs with lognormal and uniform random diffusion coefficients delivering higher-order convergence

08:30

iCal
Ana Djurdjevac | TU Berlin | Germany

Daniel Tartakovsky | Stanford University | United States

Fernando Henriquez | ETH Zurich | Switzerland

Helmut Harbrecht | University of Basel | Switzerland

MS461: Shape uncertainty quantification and applications (Part I of III)

Room:
MW HS 0250

Topic:
Uncertainty propagation

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Organized in co-operation with

GAMM AG UQ

Partial differential equations are a versatile tool to model and

eventually simulate physical phenomena. An important aspect in view

of the reliability and relevance of such simulations are uncertainties

arising from unknown parameters and measurement errors. In particular,

the modelling and discretization of uncertainties of the computational

domain requires special care. Such uncertainties emerge in a natural

fashion when considering products fabricated by line production which

are subject to manufacturing tolerances or shapes which are obtained by

remote sensing techniques, like e.g. ultrasound or magnetic resonance imaging.

This minisymposium is dedicated to recent developments in the numerical

treatment of shape uncertainties in partial differential equations

and welcomes contributions addressing analytical aspects,

forward modelling, assimilation of measurement data,

optimization, and applications.

08:30

Linear parabolic PDEs in uncertain non-cylindrical domains

09:00

- CANCELED - Stokes flow in a channel with randomly rough walls

09:30

Higher Order Quasi-Monte Carlo for the Computation of Far Field Statistics in Acoustic Wave Scattering by Uncertain Penetrable Domains

10:00

Rapid computation of far-field statistics for random obstacle scattering

08:30

iCal
Lorenzo Pareschi | University of Ferrara | Italy

Camilla Fiorini | Sorbonne Université | France

Gaël Poëtte | Atomic Energy and Alternative Energies Commission | Italy

Mattia Zanella | Politecnico di Torino | Italy

MS551: Advances in uncertainty quantification for kinetic and transport phenomena (Part I of III)

Room:
MW HS 1250

Topic:
Multiscale UQ methods

Form of presentation:
Mini-symposium

Duration:
120 Minutes

In this session we concentrate on the latest research insights for uncertainty quantification in transport problems and high-dimensional systems under structural uncertainties, with focus on kinetic and hyperbolic PDEs and multiscale interacting particle systems.

08:30

Particle based stochastic Galerkin methods for kinetic equations with random inputs

09:00

Sensitivity equation method for the Euler equations applied to uncertainty quantification

09:30

A new MC scheme for the resolution of intrusive-gPC based reduced models for the uncertain linear Boltzmann equation

10:00

Structure preserving gPC schemes for kinetic equations

08:30

iCal
Evangelos Evangelou | University of Bath | United Kingdom

Lewis Marsh | Oxford University | United Kingdom

Vasileios Maroulas | University of Tennessee | United States

Christopher Oballe | University of Tennessee | United States

MS861: Uncertainty in Data and Applications (Part I of II)

Room:
MW 0608m

Topic:
High-dimensional approximation

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Data driven discovery is the modern trend of science. A plethora of developed models are dedicated in analyzing or assimilating data arising from problems in material science and chemistry to national defense and health. The proposed mini-symposium will focus on the uncertainty of data, and its speakers will discuss techniques of uncertainty quantification, parameter estimation and noise in complex data so that robust, reproducible and convergent results are propagated. By the same token, audience and speakers will benefit from a dynamic set of prominent and auspicious speakers with heterogeneous backgrounds spanning almost the entire spectrum of mathematical sciences, from topology and geometry to statistics and machine learning.

08:30

On the choice of importance distributions for multiple importance sampling estimators

09:00

Geometric and Topological Data Analysis of Enzyme Kinetics

09:30

A Bayesian Framework for Persistent Homology

10:00

Bayesian Inference using the Pullback of a Persistence Map

08:30

iCal
Sebastian Reich | University of Potsdam | Germany

Carsten Hartmann | BTU Cottbus - Senftenberg | Germany

Abhishek Halder | University of California at Santa Cruz | United States

Simone Carlo Surace | University of Bern | Switzerland

MS721: Optimal Control Methods for Data Assimilation and Simulation

Chair(s)
Prashant Mehta (University of Illinois at Urbana-Champaign)

Nan Chen (University of Wisconsin at Madison)

Nan Chen (University of Wisconsin at Madison)

Room:
MW ZS 1050

Topic:
Data assimilation

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Duality between data assimilation/nonlinear filtering and optimal control has a rich history tracing back to Kalman-Bucy’s original 1961 paper. Duality is manifested in many guises, e.g., with the time arrow reversed, the Riccati equation of optimal control is the same as equation for the covariance update equation of the Kalman filter. In recent years, the duality relationship has been used to derive control type algorithms for data assimilation and simulation problems. This has led to several new classes of control-type algorithms such as (i) nonlinear smoothers based on approximate solution of the Bellman’s equation of optimal control; (ii) forward-backward algorithms based on a Schrodinger bridge-type construction; (iii) feedback particle filter based on a diffusion map approximation of the solution of a certain Poisson equation; and (iv) gradient flow type interpretations of linear and nonlinear filters. In numerical evaluations, it is often found that these control algorithms exhibit smaller simulation variance and better scaling properties with problem dimension when compared to the traditional methods based on importance sampling.

This session will serve to provide a snapshot of some of the exciting news developments in this historically significant area.

08:30

A Schroedinger perspective on data assimilaton

09:00

Adaptive simulation of rare events in high dimensions: a stochastic control approach

09:30

Wasserstein gradient flow for filtering and control: theory and algorithms

10:00

Feedback particle filters on manifolds for point process observations

08:30

iCal
Jinglai Li | University of Liverpool | United Kingdom

Xiaoliang Wan | Louisiana State University | United States

Zhiwen Zhang | University of Hong Kong | Hong Kong

Ke Li | ShanghaiTech University | China

MS741: Machine learning methods for reliability analysis and risk assessment (Part I of II)

Room:
MW ZS 2050

Topic:
Rare events and Risk

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Reliability analysis and risk assessment for complex physical and

engineering systems governed by partial differential equations (PDEs)

are computationally intensive, especially when high-dimensional random

parameters are involved. Since standard numerical schemes for solving

these complex PDEs are expensive, traditional Monte Carlo methods

which require repeatedly solving PDEs are infeasible. Alternative

approaches which are typically the surrogate based methods suffer from

the so-called ``curse of dimensionality'', which limits their

application to problems with high-dimensional parameters. The purpose

of this mini-symposium is to bring researchers from different fields

to discuss the recent machine learning methods for such problems,

focusing on both novel machine learning surrogates and alternative

Monte Carlo methods.

08:30

A modified Multicanonical Monte Carlo method for failure probability estimation

09:00

- CANCELED - Coupling the reduced-order model and the generative model for an importance sampling estimator

09:30

- CANCELED - A model reduction method for multiscale elliptic PDEs with random coefficients using an optimization approach

10:00

- CANCELED - A Hierarchical Neural Hybrid Method for Failure Probability Estimation

08:30

iCal
Ryan McClarren | University of Notre Dame | United States

Alexander Dowling | University of Notre Dame | United States

Dimitrios Loukrezis | Technische Universität Darmstadt | Germany

Pengfei Wei | Northwestern Polytechnical University | China

MS851: Computational Uncertainty Quantification for Energy and Power Systems

Room:
MW ZS 1450

Topic:
UQ for complex systems

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Organized in co-operation with

GAMM AG UQ

Uncertainty quantification plays a significant role in computational energy research. For instance, physical models of new energy storage are helpful tools in electromobility applications but may suffer from random inputs. Other relevant examples are renewable energy units and reliable energy network systems under volatile sources. The minisymposium covers the broad field of computational methods for energy and power systems with a particular focus on efficient methods for uncertainty quantification and sensitivity analysis. The primary purpose is to identify common methodologies and interfaces. Contributions will address applications of current interest as well as efficient algorithms and their mathematical background.

08:30

Robust Intrusive Uncertainty quantification for coupled flow physics

09:00

Probabilistic Forecasting and Stochastic Programming for Optimal Bidding in Energy Markets

09:30

Global Sensitivity Analysis of the Transient Thermal Impedance of Power Semiconductor Heat Sinks

10:00

- CANCELED - Statistical inference and stochastic simulation for uncertainty quantification with imperfect dataset

08:30

iCal
Julie Bessac | Argonne National Laboratory | United States

Nathan Urban | Los Alamos National Laboratory | United States

Timothy Palmer | University of Oxford | United Kingdom

MS451: Subgrid variability modeling and stochastic parameterization for multiscale uncertainty quantification

Chair(s)
Julie Bessac (Argonne National Laboratory)

Ahmed Attia (Argonne National Laboratory)

Emil Constantinescu (Argonne National Laboratory)

Ahmed Attia (Argonne National Laboratory)

Emil Constantinescu (Argonne National Laboratory)

Room:
MW ZS 1550

Topic:
Statistical methods for UQ

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Analysis and modeling under uncertainty are increasingly critical for robust scientific simulations. Physics-based model simulations cannot resolve the mathematical model exactly, typically leaving out fine scales, which are either approximated or not represented. This results in uncertainties in their outputs that need to be characterized. A variety of stochastic methods have been developed to address these errors and uncertainty in order to better describe complex systems. In this symposium we discuss new developments in sub-grid stochastic models, multiscale aspects, model reduction techniques, and the effect they have on Bayesian inversion and data assimilation applications.

08:30

Statistical space-time characterization of sub-grid air-sea exchanges variability including scale information

09:00

Machine learning for sub-grid parameterizations in Earth system and hydrodynamic simulations: Stochastic parameterization, uncertainty quantification, and online learning

10:00

- CANCELED - Towards A Probabilistic Earth-system model

08:30

iCal
Christopher Albert | Max-Planck-Institut für Plasmaphysik | Germany

Ulrich Callies | Helmholtz-Zentrum Geesthacht | Germany

Katharina Rath | Ludwig-Maximilians-Universität München | Germany

Sascha Ranftl | Technische Universität Graz | Austria

MS841: Uncertainty quantification for model complexity reduction

Chair(s)
Udo von Toussaint (Max Planck Institute for Plasma Physics)

Christopher Albert (Max Planck Institute for Plasma Physics)

Christopher Albert (Max Planck Institute for Plasma Physics)

Room:
MW 2250

Topic:
UQ for complex systems

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Increasingly refined numerical models that depend on a large number of parameters introduce challenges with respect to the computability and interpretability of the generated data. Methods of uncertainty quantification and sensitivity analysis offer ways to identify relevant parameters for the construction of reduced complexity models. In a well-defined parameter range this ultimately allows to replace the original model by a possibly probabilistic fast surrogate and can provide insight into the main dependencies within their range of uncertainty. The present session focuses on the development and application of such techniques that are useful for a variety of model classes from different fields. In particular this includes Bayesian methods and the interplay of uncertainty quantification with surrogates and low-fidelity models as well as methods from machine learning. Application cases ranging from environmental science and biomechanics to plasma physics will demonstrate features and limitations. They will also show similarities and differences to be taken into account for techniques that aim to be widely applicable.

08:30

profit: A framework for parameter studies via surrogate models with nested uncertainty quantification

09:00

Combining MCMC, profit and Bayesian networks to identify processes that contribute to simulated algae growth along the Elbe River

09:30

- NEW - Uncertainty quantification of Hamiltonian maps using polynomial chaos expansion

10:00

Fully Bayesian multi-fidelity uncertainty quantification with Gaussian processes applied to computational fluid dynamics of the human aorta

08:30

iCal
Areski Cousin | Universite de Strasbourg | France

Clément Rey | Ecole Polytechnique | France

Linda Chamakh | BNP Paribas | France

Cyril Benezet | Ecole Polytechnique | France

MS301: UQ in Finance

Room:
MW 1701

Topic:
UQ for complex systems

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Quantifying uncertainty in finance is a major concern when one has to address properly risk management issues with uncertainty with respect to the model and to its inputs, for instance.

In this session, we have selected different up-to-date contributions: how to derive a metamodel in credit risk, where a direct sampling of the loss is quite time-consuming (because of large number of obligors)? how to use to cleverly interpolate financial quantities (interest rate curve, implied volatility surface) with kriging techniques accounting for arbitrage-free conditions? how to account for uncertain model (with a prior distribution on the copula dependency) to compute extreme losses in capital allocation problems? or to address asset management problems (portfolio optimisations)?

The tools cover Gaussian processes, MCMC, splitting, Polynomial Chaos Expansion, Stochastic Approximations.

08:30

Kriging for arbitrage-free construction of financial term-structures

09:00

Meta-model of a large credit risk portfolio in the Gaussian copula model

09:30

Uncertain Quantification Stochastic Algorithm as a Decision-Making tool for Portfolio Optimisation

10:00

Risk measures of a mixture model, an approach with mixture of MCMC and splitting

08:30

iCal
James Salter | University of Exeter | United Kingdom

Michael Goldstein | Durham University | United Kingdom

Samuel Coveney | University of Sheffield | United Kingdom

Victoria Volodina | Alan Turing Institute | United Kingdom

MS711: Inverse modelling using History Matching (Part I of II)

Room:
MW HS 0337

Topic:
Model error, discrepancy and calibration

Form of presentation:
Mini-symposium

Duration:
120 Minutes

[ Moved from MW HS 2235 ]

History matching is a way of inverse modelling, or calibrating/tuning, the inputs of a complex numerical model given observations on the outputs. History matching is very different to ways of performing a Bayesian calibration. For example, the result is not a posterior distribution on the model inputs, but a set of model input points that are not implausible points given the data. It is not probabilistic. The idea is simple. A series of waves of model runs is carried out. At each wave the scaled distance (the implausibility measure) between the observations and the expected value of an emulator (either a Gaussian or a second order process) of the model for all inputs is calculated. If this distance is too large the set of inputs is ruled implausible. The scaling consists of three components:- the emulator variance (known but input dependent), the observation variance (so poor observations are downweighted compared to more accurate ones) and a variance that measures the discrepancy between the model and the real world. After the first wave a new wave of model runs is carried out in the Not Ruled Out Yet (NROY) space. A new emulator is derived and new implausibilities calculated. At each wave the emulator becomes a better fit to the model so the NROY space is progressively reduced. This mini-symposium is concerned with the application and extension of history matching to a variety of applications.

08:30

Efficient calibration for spatio-temporal models using basis methods

09:00

- CANCELED - History matching by implausibility emulation

09:30

Constructing personalised tissue parameter maps of human left atrium from clinical measurements

10:00

Bayesian Optimal Design for iterative refocussing

08:30

iCal
Saddam N Y Hijazi | SISSA Trieste | Italy

Bassel Saleh | Oden Institute for Computational Engineering and Sciences, The University of Texas at Austin | United States

Davide Torlo | University of Zurich | Switzerland

MS351: Reduced order methods for uncertainty quantification in CFD parametric problems

Chair(s)
Gianluigi Rozza (SISSA Trieste)

Giovanni Stabile (SISSA Trieste)

Francesco Ballarin (SISSA Trieste)

Giovanni Stabile (SISSA Trieste)

Francesco Ballarin (SISSA Trieste)

Room:
IAS 0.001

Topic:
Reduced order models

Form of presentation:
Mini-symposium

Duration:
120 Minutes

The mini deals with state of the art reduced order methods for uncertainty quantification in parametric computational fluid dynamics (CFD) problems dealing with data assimilation, data reconstruction, random inputs. Special attention is devoted to inverse problems for optimisation and control, as well as to nonlinear problems. Complex applications of the methodology are considered in industrial setting, as well as in medicine.

08:30

- NEW - Non-Intrusive Polynomial Chaos Method Applied to Full-Order and Reduced Problems in Computational Fluid Dynamics

09:30

Neural Networks as Control Variates for UQ in Ice Sheet Flow

10:00

Weighted model order reduction techniques for advection dominated problems with selective stabilization

08:30

iCal
Roman Sueur | EDF R&D | France

Joseph Hart | Sandia National Laboratories | United States

Guillaume Perrin | CEA/DAM | France

Amandine Marrel | CEA/DEN | France

MS131: Robustness analysis of UQ to distribution uncertainty (Part I of II)

Room:
IAS 4.001

Topic:
Rare events and Risk

Form of presentation:
Mini-symposium

Duration:
120 Minutes

One of the most critical hypothesis in uncertainty quantification studies is the choice of the distributions of uncertain input variables which are propagated through the numerical model. In general, such pdf come from various sources (statistical inference, design or operation rules, expert judgment, calibration, etc.), and are then established with a certain level of accuracy or confidence. Moreover, in many applications, related for example to industrial safety, engineers are not able to assign a given probability distribution to some of the inputs. This happens for example for inputs corresponding to physical parameters for which no data are available.

Hence, bringing stringent justifications to the overall approach requires quantifying the impact of the pdf modeling assumptions on the quantity of interest (QoI). In this context, the “input pdf robustness analysis”, has been recently defined as a particular setting of the sensitivity analysis domain (like the screening one or the quantitative partitioning one). Various QoI can be considered, as the mean of the model output, its variance, a probability that the output exceeds a threshold, a quantile of the output or even sensitivity indices.

This Minisymposium, which will be held in two parts (4 presentations in each part), aims at presenting several recent theoretical developments on this subject, as well as practical and industrial issues.

08:30

An informative law perturbation approach in Robustness Analysis

09:00

Robustness of Sobol’ indices to distributional uncertainty

09:30

Quantification of the impact of an imprecise specification of input distributions on reliability analysis results

10:00

Sensitivity analysis with dependence measures under uncertainty of input distribution

08:30

iCal
Andrew Gordon Wilson | NYU Courant | United States

Siddharth Swaroop | University of Cambridge | United Kingdom

Yang Song | Stanford University | United States

Nikola Kovachki | California Institute of Technology | United States

MS691: Uncertainty Quantification in Deep Learning (Part I of III)

Chair(s)
Paris Perdikaris (University of Pennsylvania)

Phaedon-Stelios Koutsourelakis (Technical University of Munich)

Phaedon-Stelios Koutsourelakis (Technical University of Munich)

Room:
Interims Lecture Hall 101

Topic:
Physics-informed and data-driven UQ methods

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Organized in co-operation with

GAMM AG UQ

Deep learning techniques are becoming the center of attention across many scientific disciplines. Many predictive tasks are currently being tackled using over-parameterized, black-box discriminative models such as deep neural networks, in which interpretability and robustness is often sacrificed in favor of flexibility in representation and scalability in computation. Such models have yielded remarkable results in data-rich domains, yet their effectiveness in data-scarce and risk-sensitive tasks still remains questionable, primarily due to open challenges in statistical inference and uncertainty quantification. This mini-symposium invites contributions on uncertainty quantification methods for deep learning and their application in the physical and engineering sciences. Topics include (but are not limited to) Bayesian neural networks, deep generative models, posterior inference techniques, and applications to forward/inverse problems, active learning, Bayesian optimization and reinforcement learning.

08:30

Subspace Inference for Bayesian Deep Learning

09:00

Applying Bayesian Principles to Deep Learning: Scaling, Uncertainty Calibration, and Continual Learning

09:30

Analyzing the Vulnerability of Deep Classifiers with Deep Generative Models

10:00

Model Reduction for Input-Output Maps

08:30

iCal
Edoardo Patelli | Strathclyde University | United Kingdom

Bert Debusschere | Sandia National Laboratories | United States

Tillmann Mühlpfordt | Karlsruhe Institute of Technology | Germany

Aakash Bangalore Satish | Johns Hopkins University | United States

MS821: Software for UQ (Part I of III)

Chair(s)
Tobias Neckel (Technical University of Munich)

Dirk Pflüger (University of Stuttgart)

Stefano Marelli (ETH Zurich)

Edoardo Patelli (Strathclyde University)

Dirk Pflüger (University of Stuttgart)

Stefano Marelli (ETH Zurich)

Edoardo Patelli (Strathclyde University)

Room:
Interims Lecture Hall 102

Topic:
Other

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Organized in co-operation with

GAMM AG UQ

With the ever increasing importance of UQ in various disciplines and fields, software solutions and libraries for UQ problems get more and more important. Progress and use of UQ techniques relies on the availability of software features and support. This raises interesting questions for the UQ community such as: What are the current properties of available tools? For which classes of problems have they been developed? What methods or algorithms do they provide? What are challenges for UQ software and which resources are required? What are recent improvements? What are the next steps and the long-term goals of the development?

This minisymposium brings together experts for different software in the context of UQ, ranging from tools that ease up individual tasks of UQ (such as surrogate modelling, UQ workflows, dimensionality reduction, data augmentation) up to whole frameworks for solving UQ problems. The minisymposium will foster discussion and exchange of ideas between developers and (prospective) users.

08:30

Cossan Software: Efficient and user-friendly computational tools for dealing with uncertainty

09:00

- CANCELED - UQTk, a C++/Python Toolkit for Uncertainty Quantification: Overview and Applications

09:30

PolyChaos.jl – An Open Source Julia Package for Orthogonal Polynomials, Quadrature, and Polynomial Chaos Expansion

10:00

UQpy: A Python toolkit and development environment for UQ

08:30

iCal
David Sirl | University of Nottingham | United Kingdom

Peter Simon | Eötvös Loránd University | Hungary

Tiago Peixoto | Central European University | Hungary

Jean-Charles Croix | University of Sussex | United Kingdom

MS391: Stochastic processes on large-scale Networks

Room:
Exzellenzzentrum 0003

Topic:
Statistical methods for UQ

Form of presentation:
Mini-symposium

Duration:
120 Minutes

The behaviour of many large-scale systems can be modelled by a network of pairwise interaction. Examples include spread of epidemics, neural activity in the brain and social media influence. While these models are relatively flexible, their complexity strongly depends both on the structural properties of networks and the precise nature of the process unfolding on them. The complete specification of such models require various amount and quality of information. In the majority of situations, however, such data sets are incomplete and contain errors. Furthermore, pairwise interactions on networks in many instances are hidden from us and are impossible or very difficult to measure directly. Problems of interest in such situations include quantifying the effect of errors and omissions in the data on the predictability of the behaviour of the process unfolding on large-scale networks, and the inference of the underlying structure from partial, erroneous and usually indirect observations. In this symposium we consider different approaches to such problems. This includes approximation of large scale networks through statistical averaging techniques and Bayesian inference of the network structure.

08:30

A network SIR epidemic model with preventive rewiring

09:00

Comparison of different statistical averaging methods for epidemic processes on networks

09:30

- NEW - Network reconstruction and community detection from dynamics

10:00

Non-parametric Bayesian inference for density dependent Networks with SIS discrete data

11:00

iCal
Anthony Nouy | Centrale Nantes / LMJL | France

IP04: Anthony Nouy: Approximation and Learning with Tree Tensor Networks

Room:
MW HS 2001

Topic:
High-dimensional approximation

Form of presentation:
Plenary Lecture

Duration:
45 Minutes

11:00

Approximation and Learning with Tree Tensor Networks

11:00

iCal

IP04 - streamed from HS 2001: Anthony Nouy: Approximation and Learning with Tree Tensor Networks

Room:
MW HS 0001

Topic:
High-dimensional approximation

Form of presentation:
Plenary Lecture

Duration:
45 Minutes

11:45

iCal
Elaine Spiller | Marquette University | United States

IP05: Elaine Spiller: Assessing and forecasting hazards in an uncertain future

Room:
MW HS 2001

Topic:
UQ for complex systems

Form of presentation:
Plenary Lecture

Duration:
45 Minutes

11:45

Assessing and forecasting hazards in an uncertain future

11:45

iCal

IP05 - streamed from HS 2001: Elaine Spiller: Assessing and forecasting hazards in an uncertain future

Room:
MW HS 0001

Topic:
UQ for complex systems

Form of presentation:
Plenary Lecture

Duration:
45 Minutes

14:00

iCal
Ahmed Attia | Argonne National Laboratory | United States

Dariusz Uciński | Institute of Control and Computation Engineering, University of Zielona Góra | Poland

Elizabeth Herman | Department of Mathematics, North Carolina State University | United States

Keyi Wu | University of Texas at Austin | United States

MS011: Recent advances and challenges in optimal experimental design for large-scale inverse problems (Part I of II)

Chair(s)
Ahmed Attia (Argonne National Laboratory)

Alen Alexanderian (North Carolina State University)

Alen Alexanderian (North Carolina State University)

Room:
MW HS 2001

Topic:
Design of experiments

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Computer models play an essential role in forecasting complicated phenomena such as the atmosphere, ocean dynamics, seismology among others. These models, however, are typically imperfect due to various sources of uncertainty. Measurements are snapshots of reality that are collected as an additional source of information and are used to update and even correct the model-based simulations or forecasts. The accuracy of the overall simulations and model-based forecasts is greatly influenced by the quality of the observational grid design used to collect measurements. Optimal data acquisition can be formulated as an optimal experimental design (OED) problem. The framework of model-based OED has gained wide popularity and attention from researchers in various fields in statistics, engineering, applied math and others. Challenges in model-based OED include high-dimensionality, misrepresentation of prior knowledge, increasing deviation from Gaussianity, high correlations of spatiotemporal observations, among others. This minisymposium aims to showcase the latest developments in tackling the challenges in the field of model-based OED for large-scale inverse problems.

14:00

Goal-Oriented Optimal Experimental Design Framework for Sensor Placement and Acquisition of Highly-Correlated Data

14:30

Majorization-minimization algorithm for D-optimal sensor selection in the presence of correlated measurement noise

15:00

Computing A-optimal design of experiments for large-scale inverse problems using randomized methods and reweighted $\ell_1$ minimization

15:30

- MOVED from MS012 - A Stein Variational Newton Method for Optimal Experimental Design Problems

14:00

iCal
Olivier Zahm | Inria | France

Tommaso Taddei | INRIA | France

Iason Papaioannou | Engineering Risk Analysis Group, Technische Universität München | Germany

Gabriele Santin | Center for Information and Communication Technology, Fondazione Bruno Kessler | Italy

MS671: Achieving a data-model synergy in UQ (Part I of II)

Room:
MW HS 0001

Topic:
Reduced order models

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Recent years have seen the flourishing of techniques devoted to best incorporate data in the models, either for the solution of inverse problems or for approximation purposes. This includes domain-aware Machine Learning techniques, dynamic mode decomposition or data driven model order reduction methods. This minisymposium aims to provide a venue for young researchers focusing on the theoretical analysis, the development and the application of these methodologies.

14:00

- NEW - A data free likelihood-informed subspace for dimensionality reduction of Bayesian inverse problems

14:30

A registration method for Model Order Reduction: data compression and geometry reduction

15:00

PLS-based dimension reduction for uncertainty quantification

15:30

Kernel-based surrogate models for UQ

14:00

iCal
Maggie Johnson | Jet Propulsion Laboratory, California Institute of Technology | United States

Pratik Patil | Carnegie Mellon University | United States

Kerry Cawse-Nicholson | Jet Propulsion Laboratory, California Institute of Technology | United States

Otto Lamminpää | Finnish Meteorological Institute | Finland

MS442: Uncertainty Quantification for Earth Remote Sensing (Part II of III)

Room:
MW HS 1801

Topic:
Inverse problems

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Numerous Earth-observing satellites provide high-resolution and high-volume data that facilitate scientific inference on physical and environmental processes. Most remote sensing data products used for scientific investigations are often subject to multiple stages of processing before they are widely used, and the scientific utility of these data products critically depends on a comprehensive assessment of the sources of uncertainty encountered in these stages of processing. One key stage involves the use of a retrieval algorithm to infer a geophysical quantity of interest from a satellite’s observed intensity of radiation.

The retrieval is an inverse problem that has been implemented mathematically and computationally in numerous ways for different satellite missions. Several of the presentations in this mini-symposium will each highlight an individual Earth-observing satellite and its retrieval methodology, emphasizing important contributions to uncertainty in retrieval data products. Methodological developments that interrogate the joint distribution of true geophysical states, retrieved states, and observed satellite spectra will be introduced. The presentations will span multiple Earth science applications, including weather and climate, the carbon cycle, air quality, atmospheric chemistry, and ecosystem health.

14:00

Uncertainty quantification for NASA’s Microwave Limb Sounder

14:30

Objective frequentist uncertainty quantification for atmospheric CO2 retrieval

15:00

Applications of spatial-statistical models to UQ for Earth observing missions: ECOSTRESS and SBG

15:30

Accelerated MCMC for remote sensing of atmospheric CO2

14:00

iCal
Elnaz Esmaeilzadeh Seylabi | University of Nevada | United States

Ilona Ambartsumyan | University of Texas at Austin | United States

Jinlong Wu | California Institute of Technology | United States

Oliver Dunbar | California Institute of Technology | United States

MS411: Bayesian Inverse Problems and Experimental Design for Complex Systems

Chair(s)
Jinlong Wu (California Institute of Technology)

Oliver Dunbar (California Institute of Technology)

Oliver Dunbar (California Institute of Technology)

Room:
MW HS 0350

Topic:
Inverse problems

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Bayesian inverse problems of complex systems are usually intractable, mainly due to the expansive forward simulations of the system. In practice, both derivative-free methods (e.g., ensemble Kalman inversion) and fast adjoint methods serve as promising candidates of optimization scheme to approximately solve Bayesian inverse problems for complex systems. In this session, we include talks of solving inverse problems of complex systems by using either ensemble Kalman methods or fast adjoint method. With the forward simulations evaluated in the optimization scheme, it is possible to further build surrogate models for MCMC. In several talks of this session, physics-informed approaches are also discussed in the context of Bayesian inverse problems. On the other hand, high-fidelity data of complex systems are expensive to simulate or measure, making experimental design process critical in order to obtain the most information from the system under limited resources. This experimental design topic is also discussed in this session.

14:00

- CANCELED - Near surface site characterization using ensemble Kalman inversion

14:30

Fast methods for Bayesian inverse problems governed by random PDE forward models

15:00

Estimating model-form uncertainty for multi-scale systems

15:30

Bayesian optimal design to target simulation within an idealized climate model

14:00

iCal
Carlos Jerez-Hanckes | Universidad Adolfo Ibáñez | Chile

Ulrich Römer | TU Braunschweig | Germany

Alexander Litvinenko | RWTH Aachen | Germany

Jürgen Dölz | TU Darmstadt | Germany

MS462: Shape uncertainty quantification and applications (Part II of III)

Room:
MW HS 0250

Topic:
Uncertainty propagation

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Organized in co-operation with

GAMM AG UQ

Partial differential equations are a versatile tool to model and eventually simulate physical phenomena. An important aspect in view of the reliability and relevance of such simulations are uncertainties arising from unknown parameters and measurement errors. In particular, the modelling and discretization of uncertainties of the computational domain requires special care. Such uncertainties emerge in a natural fashion when considering products fabricated by line production which are subject to manufacturing tolerances or shapes which are obtained by remote sensing techniques, like e.g. ultrasound or magnetic resonance imaging. This minisymposium is dedicated to recent developments in the numerical treatment of shape uncertainties in partial differential equations and welcomes contributions addressing analytical aspects, forward modelling, assimilation of measurement data, optimization, and applications.

14:00

Domain Uncertainty Quantification in Computational Electromagnetics

14:30

Shape Uncertainty Quantification in Computational Nano-Optics

15:00

Computation of Electromagnetic Fields Scattered From Objects of Uncertain Shapes Using Multilevel Monte Carlo

15:30

A higher order perturbation approach for electromagnetic scattering problems on random domains

14:00

iCal
Andrea Tosin | Politecnico di Torino | Italy

Jonas Kusch | Karlsruher Institut für Technologie | Germany

Anjali Nair | University of Wisconsin-Madison | United States

MS552: Advances in uncertainty quantification for kinetic and transport phenomena (Part II of III)

Room:
MW HS 1250

Topic:
Multiscale UQ methods

Form of presentation:
Mini-symposium

Duration:
120 Minutes

In this session we concentrate on the latest research insights for uncertainty quantification in transport problems and high-dimensional systems under structural uncertainties, with focus on kinetic and hyperbolic PDEs and multiscale interacting particle systems.

14:00

Multi-agent systems with uncertain interactions: the case of vehicular traffic

14:30

Intrusive acceleration techniques for uncertainty quantification

15:00

Reconstructing phonon transmission coefficients at solid interfaces using metrology

14:00

iCal
Clement Etienam | University of Manchester | United Kingdom

Alessandro Lanteri | University of Turin | Italy

Anna Little | Michigan State University | United States

James Murphy | Tufts University | United States

MS862: Uncertainty in Data and Applications (Part II of II)

Room:
MW 0608m

Topic:
High-dimensional approximation

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Data driven discovery is the modern trend of science. A plethora of developed models are dedicated in analyzing or assimilating data arising from problems in material science and chemistry to national defense and health. The proposed mini-symposium will focus on the uncertainty of data, and its speakers will discuss techniques of uncertainty quantification, parameter estimation and noise in complex data so that robust, reproducible and convergent results are propagated. By the same token, audience and speakers will benefit from a dynamic set of prominent and auspicious speakers with heterogeneous backgrounds spanning almost the entire spectrum of mathematical sciences, from topology and geometry to statistics and machine learning.

14:00

An Ultra-fast procedure for Learning Discontinuous Functions

14:30

A Bayesian Nonparametric Model for Longitudinal Data in Sport

15:00

Wavelet invariants for statistically robust multi-reference alignment

15:30

Diffusion Geometric Approaches to Active Learning

14:00

iCal
Andrey A Popov | Virginia Tech | United States

Javier Amezcua | University of Reading | United Kingdom

Elias D. Nino-Ruiz | Universidad del Norte | Colombia

Paul J. Rozdeba | University of Potsdam | Germany

MS091: Algorithms for Large Scale and Non-linear Data Assimilation (Part I of II)

Room:
MW ZS 1050

Topic:
Data assimilation

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Accuracy is always at odds with efficiency in the context of Data Assimilation on complex dynamical system. Such systems often involve large amounts of variables, with impactful non-linearities, and poorly understood stochastic behaviour. Tackling these problems in an efficient manner is the key to unlocking the next generation of algorithms. The discussion of directions such as exploiting the time-dependent structure of natural systems, reduced order modeling, accounting for model error, and efficient ways to solve the underlying optimization problem, are just some of the topics of fundamental importance in the next few years of research, that will be covered.

14:00

Data-driven Reduced Order Model Control Variates in a Multilevel Ensemble Kalman Filter

14:30

Model error with time-autocorrelation and its effect in data assimilation

15:00

- NEW - A Data-Driven Localization Method for Ensemble Based Data Assimilation

15:30

Ensemble methods for nonparametric drift estimation and model selection

14:00

iCal
Xiang Zhou | City University of Hong Kong | Hong Kong

Kejun Tang | ShanghaiTech University | China

Dingjiong Ma | University of Hong Kong | Hong Kong

Qifeng Liao | ShanghaiTech University | China

MS742: Machine learning methods for reliability analysis and risk assessment (Part II of II)

Room:
MW ZS 2050

Topic:
Rare events and Risk

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Reliability analysis and risk assessment for complex physical and

engineering systems governed by partial differential equations (PDEs)

are computationally intensive, especially when high-dimensional random

parameters are involved. Since standard numerical schemes for solving

these complex PDEs are expensive, traditional Monte Carlo methods

which require repeatedly solving PDEs are infeasible. Alternative

approaches which are typically the surrogate based methods suffer from

the so-called ``curse of dimensionality'', which limits their

application to problems with high-dimensional parameters. The purpose

of this mini-symposium is to bring researchers from different fields

to discuss the recent machine learning methods for such problems,

focusing on both novel machine learning surrogates and alternative

Monte Carlo methods.

14:00

Generative Model based on Winslow Mapping for Sampling unnormalised Distribution

14:30

- CANCELED - Deep density estimation for Fokker-Planck equations using flow-based generative model

15:00

Proper orthogonal decomposition method for multiscale elliptic PDEs with random coefficients

15:30

- CANCELED - Domain decomposed uncertainty analysis based on RealNVP

14:00

iCal
Matthias Chung | Technical University of Berlin | Germany

Sebastian Rojas Gonzalez | KU Leuven | Belgium

Victor Picheny | Prowler.io | United Kingdom

MS761: Statistical Surrogate Modeling and Optimization for Stochastic Simulation (Part I of II)

Room:
MW ZS 1450

Topic:
Surrogate models

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Historically, design and analysis of computer experiments focused on deterministic solvers from the physical sciences via Gaussian process (GP) interpolation. But nowadays computer modeling is common in the social, management and biological sciences, where stochastic simulations abound. In this minisymposium, we bring together a selection of researchers in the areas of statistical surrogate modeling, active learning, and Bayesian optimization of stochastic computer model, simulation campaigns, and high volume observational studies. Noisier simulations demand bigger experiments to isolate signal from noise, and more sophisticated GP models -- such as adding a variance processes to track changes in noise throughout the input space in the face of heteroskedasticity. Appropriate surrogate modeling is key to the propagation of uncertainty to decision criteria underlying important large-scale and real time control of systems which rely on expensive simulation campaigns. Think of synthesis between off-line simulation of urban road traffic and ride demand with on-line measurements from potential riders and their routes in a rideshare pool. Or similarly the combination of limited data on disease spread combined with social-network backed simulation of epidemiological dynamics and entertainment of intervention strategies such as vaccination and quarantine. The talks will be on these methodologies and applied in those challenging modeling and optimization real-world problems.

14:00

From Parameter and Uncertainty Estimation to Optimal Experimental Design: Challenges in Biological Dynamical Systems Inference

15:00

Multiobjective stochastic simulation optimization with correlation and heterogeneous noise

15:30

Spatialised Generalized lambda distribution for risk-averse Bayesian Optimisation

14:00

iCal
Kevin Lin | University of Arizona | United States

Nisha Chandramoorthy | Massachusetts Institute of Technology | United States

Alexis-Tzianni Charalampopoulos | Massachusetts Institute of Technology | United States

Cecilia Mondaini | Drexel University | United States

MS121: Computational Statistics meets Computational Dynamics (Part I of II)

Chair(s)
Ben Zhang (Massachusetts Institute of Technology)

Tuhin Sahai (United Technologies Research Center)

Kevin Lin (University of Arizona)

Tuhin Sahai (United Technologies Research Center)

Kevin Lin (University of Arizona)

Room:
MW ZS 1550

Topic:
Statistical methods for UQ

Form of presentation:
Mini-symposium

Duration:
120 Minutes

In this minisymposium, we explore the symbiotic relationship between computational statistics and computational dynamics. The interaction of the two fields have long been established. Efficiently computing statistics of dynamical quantities is of interest in science and engineering, and cleverly constructed dynamical systems are used to sample from high-dimensional probability distributions. We will highlight recent advances in numerical methods that utilize tools in one field to solve problems in the other in a novel fashion. We will exhibit new algorithms for sensitivity analysis, efficient sampling methods, and inference.

14:00

Couplings-based sensitivity estimates for stochastic dynamics

14:30

Computing statistical response to small perturbations in chaotic systems

15:00

Machine learning non-local closures for turbulent anisotropic multiphase fluid flows

15:30

Numerical approximation for invariant measures of the 2D Navier-Stokes equations

14:00

iCal
Tim Dodwell | University of Exeter | United Kingdom

Colin Fox | University of Otago | New Zealand

Daniel Schaden | Technical University of Munich | Germany

Jordan Franks | Newcastle University | United Kingdom

MS811: Multilevel and Multi-fidelity Methods for Model-Based Statistical Learning (Part I of III)

Chair(s)
Tiangang Cui (Monash University)

Santiago Badia (Monash University)

Alexander Gilbert (Ruprecht-Karls University Heidelberg)

Youssef Marzouk (Massachusetts Institute of Technology)

Robert Scheichl (Ruprecht-Karls University Heidelberg)

Santiago Badia (Monash University)

Alexander Gilbert (Ruprecht-Karls University Heidelberg)

Youssef Marzouk (Massachusetts Institute of Technology)

Robert Scheichl (Ruprecht-Karls University Heidelberg)

Room:
MW 2250

Topic:
Statistical methods for UQ

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Organized in co-operation with

GAMM AG UQ

The synthesis of various information sources, including a priori domain knowledge, statistical assumptions, field data, etc., large-scale numerical models is one of the key steps in building interpretable and predictive models for supporting critical decisions in science, engineering, medicine, and beyond. Typical examples can be found in oil/gas reservoir modeling, treatment of saltwater intrusion, medical imaging, tumor treatment, aircraft design. Because of the computationally costly nature of the numerical models and stringent requirements on the accuracy of the statistical learning outcomes, multilevel and multi-fidelity methods provide a viable route for solving these model-based statistical learning tasks. This mini-symposium will bring together researchers working on the forefront of multilevel and multi-fidelity methods (and other relevant methods) intended to accelerate model-based statistical learning tasks.

14:00

- NEW - Data Driven Multiscale Methods for Bayesian Inverse Problems based on Large Scale Partial Differential Equations

14:30

Multilevel MCMC as a Surrogate Transition Method

15:00

Multilevel Best Linear Unbiased Estimators

15:30

- NEW - Unbiased inference for discretely observed hidden Markov model diffusions

14:00

iCal
Christian Bayer | WIAS Berlin | Germany

Paolo Pigato | University of Rome "Tor Vergata" | Italy

Mikko Pakkanen | Imperial College London | United Kingdom

Ludovic Goudenège | Fédération de Mathématiques de CentraleSupélec | France

MS171: UQ for Rough Volatility and Predictive Models in Finance (Part I of II)

Room:
MW 1701

Topic:
UQ for complex systems

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Rough volatility models are an increasingly popular class of models in quantitative finance. In contrast to conventional stochastic volatility models, the volatility is driven by a fractional Brownian motion with Hurst index H < 1/2 which is rougher than Brownian motion. This change greatly improves the fit to time series data of underlying asset prices as well as to option prices, see, for instance, [Bayer, Friz, Gatheral, Quantitative Finance 16(6), 887-904, 2016]. Hence, introducing non-Markovian noise improves the predictive power of the model while maintaining parsimoniousness. Unfortunately, the loss of the Markov property poses severe challenges for theoretical and numerical analyses as well as for computational practice.

This minisymposium brings together different approaches for various UQ tasks in the context of rough volatility models and predictive models in finance. The problems addressed range from calibration and statistical analysis of the model parameters to optimal control of rough volatility models. To overcome the considerable practical hurdles posed by the lack of Markovianity, the contributions to the minisymposium use diverse tools such as deep neural networks and large deviation theory, assisted by properly analyzed simulation techniques.

14:00

Weak Error Rates for Option Pricing under the Rough Bergomi Model

14:30

Analysis of rough volatility via rough paths / regularity structures

15:00

Moment-based estimation of log-normal volatility models

15:30

Machine Learning for Pricing American Options in High-Dimensional Markovian and non-Markovian models

14:00

iCal
Samuel Jackson | University of Southampton | United Kingdom

Wenzhe Xu | University of Exeter | United Kingdom

Evan Baker | University of Exeter | United Kingdom

Doug McNeall | Met Office and University of Exeter | United Kingdom

MS712: Inverse modelling using History Matching (Part II of II)

Room:
MW HS 0337

Topic:
Model error, discrepancy and calibration

Form of presentation:
Mini-symposium

Duration:
120 Minutes

[ Moved from MW HS 2235 ]

History matching is a way of inverse modelling, or calibrating/tuning, the inputs of a complex numerical model given observations on the outputs. History matching is very different to ways of performing a Bayesian calibration. For example, the result is not a posterior distribution on the model inputs, but a set of model input points that are not implausible points given the data. It is not probabilistic. The idea is simple. A series of waves of model runs is carried out. At each wave the scaled distance (the implausibility measure) between the observations and the expected value of an emulator (either a Gaussian or a second order process) of the model for all inputs is calculated. If this distance is too large the set of inputs is ruled implausible. The scaling consists of three components:- the emulator variance (known but input dependent), the observation variance (so poor observations are downweighted compared to more accurate ones) and a variance that measures the discrepancy between the model and the real world. After the first wave a new wave of model runs is carried out in the Not Ruled Out Yet (NROY) space. A new emulator is derived and new implausibilities calculated. At each wave the emulator becomes a better fit to the model so the NROY space is progressively reduced. This mini-symposium is concerned with the application and extension of history matching to a variety of applications.

14:00

Design of Physical Experiments for History Matching

14:30

Enhancing Uncertainty Quantification using machine learning approaches.

15:00

Future Proofing a Building Using History Matching Inspired Level Set Techniques

15:30

History matching a land surface simulator

14:00

iCal
Dirk Husmeier | University of Glasgow | United Kingdom

Richard Clayton | The University of Sheffield | United Kingdom

Martin Hess | SISSA International School for Advanced Studies | Italy

Mihaela Paun | University of Glasgow | United Kingdom

MS581: Advances in Inverse Problems and Uncertainty Quantification in Cardiovascular Modeling: (Part I of II)

Chair(s)
Mitchel Colebank (North Carolina State University)

Dirk Husmeier (University of Glasgow)

Mette Olufsen (North Carolina State University)

Dirk Husmeier (University of Glasgow)

Mette Olufsen (North Carolina State University)

Room:
IAS 0.001

Topic:
UQ for complex systems

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Advances in computational medicine have made mathematical modeling of hemodynamics a key area of scientific research. Innovations in high performance computing and high-fidelity models allows for sophisticated approximations of in-vivo cardiovascular dynamics. To this end, a variety of models including system level 0D models, 1D fluid dynamics network models, and 3D fluid structure interaction models, can be used to investigate structure-function relation of the cardiovascular system, on a local, global, or multiscale level. However, these computational models are susceptible to both model discrepancy and uncertainty in model inputs, and predictions. Cardiovascular models are calibrated to sparse data, i.e. they contain parameters unmeasurable in-vivo, making parameter estimation and forward uncertainty propagation difficult. This minisymposium will focus on cardiovascular inverse problems and statistical inference methodology including:

• Parameter estimation techniques for complex ODE-PDE coupled models

• Novel emulation and metamodeling procedures for high-fidelity models

• Advances in surrogate and low-fidelity model construction

• Quantification of model consistency using machine-learning

• Efficient uncertainty propagation and quantification

• Innovative numerical and analytical sensitivity techniques

14:00

Statistical inference in soft-tissue mechanics with an application to prognostication of myocardial infarction

14:30

Emulating cardiac cell models with Gaussian processes

15:00

State of the art and perspectives in patient-specific reduced order modelling for cardiovascular problems

15:30

Parameter estimation and uncertainty quantification for the pulmonary circulation system

14:00

iCal
Michael D. Shields | Johns Hopkins University | United States

Luc Bonnet | ONERA - University Paris-Saclay | France

Jérôme Stenger | Université Toulouse Paul Sabatier | France

Vincent Chabridon | EDF R&D | France

MS132: Robustness analysis of UQ to distribution uncertainty (Part II of II)

Room:
IAS 4.001

Topic:
Rare events and Risk

Form of presentation:
Mini-symposium

Duration:
120 Minutes

One of the most critical hypothesis in uncertainty quantification studies is the choice of the distributions of uncertain input variables which are propagated through the numerical model. In general, such pdf come from various sources (statistical inference, design or operation rules, expert judgment, calibration, etc.), and are then established with a certain level of accuracy or confidence. Moreover, in many applications, related for example to industrial safety, engineers are not able to assign a given probability distribution to some of the inputs. This happens for example for inputs corresponding to physical parameters for which no data are available.

Hence, bringing stringent justifications to the overall approach requires quantifying the impact of the pdf modeling assumptions on the quantity of interest (QoI). In this context, the “input pdf robustness analysis”, has been recently defined as a particular setting of the sensitivity analysis domain (like the screening one or the quantitative partitioning one). Various QoI can be considered, as the mean of the model output, its variance, a probability that the output exceeds a threshold, a quantile of the output or even sensitivity indices.

This Minisymposium, which will be held in two parts (4 presentations in each part), aims at presenting several recent theoretical developments on this subject, as well as practical and industrial issues.

14:00

The multi-model approach to uncertainty quantification and propagation from small data sets: An overview

14:30

Post-Optimal Design using Optimal Uncertainty Quantification

15:00

Optimal Uncertainty Quantification of a risk measurement on moment class

15:30

Reliability-oriented Sobol’ indices under distribution parameter uncertainty

14:00

iCal
Thomas O'Leary Roseberry | UT-Austin | United States

Shengyang Sun | University of Toronto | Canada

Sebastian Kaltenbach | Technical University of Munich | Germany

Maximilian Rixner | Technical University of Munich | Germany

MS692: Uncertainty Quantification in Deep Learning (Part II of III)

Chair(s)
Paris Perdikaris (University of Pennsylvania)

Phaedon-Stelios Koutsourelakis (Technical University of Munich)

Phaedon-Stelios Koutsourelakis (Technical University of Munich)

Room:
Interims Lecture Hall 101

Topic:
Physics-informed and data-driven UQ methods

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Organized in co-operation with

GAMM AG UQ

Deep learning techniques are becoming the center of attention across many scientific disciplines. Many predictive tasks are currently being tackled using over-parameterized, black-box discriminative models such as deep neural networks, in which interpretability and robustness is often sacrificed in favor of flexibility in representation and scalability in computation. Such models have yielded remarkable results in data-rich domains, yet their effectiveness in data-scarce and risk-sensitive tasks still remains questionable, primarily due to open challenges in statistical inference and uncertainty quantification. This mini-symposium invites contributions on uncertainty quantification methods for deep learning and their application in the physical and engineering sciences. Topics include (but are not limited to) Bayesian neural networks, deep generative models, posterior inference techniques, and applications to forward/inverse problems, active learning, Bayesian optimization and reinforcement learning.

14:00

- Moved from CT16 - Encoder-decoder architectures for PDE based inference problems

14:30

Benchmarking and Exploiting Predictive Correlations in Deep Learning

15:00

- Moved from CT14 - Deep probabilistic learning of reduced dynamics of multiscale systems in the Small Data regime

15:30

Deep probabilistic reduced-order models accounting for physics by using virtual observables

14:00

iCal
Miroslav Stoyanov | Oak Ridge National Laboratory | United States

Jonathan Feinberg | Expert Analytics | Norway

Florian Künzner | Technical University of Munich | Germany

Simen Tennøe | Expert Analytics | Norway

MS822: Software for UQ (Part II of III)

Chair(s)
Tobias Neckel (Technical University of Munich)

Dirk Pflüger (University of Stuttgart)

Stefano Marelli (ETH Zurich)

Edoardo Patelli (Strathclyde University)

Dirk Pflüger (University of Stuttgart)

Stefano Marelli (ETH Zurich)

Edoardo Patelli (Strathclyde University)

Room:
Interims Lecture Hall 102

Topic:
Other

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Organized in co-operation with

GAMM AG UQ

With the ever increasing importance of UQ in various disciplines and fields, software solutions and libraries for UQ problems get more and more important. Progress and use of UQ techniques relies on the availability of software features and support. This raises interesting questions for the UQ community such as: What are the current properties of available tools? For which classes of problems have they been developed? What methods or algorithms do they provide? What are challenges for UQ software and which resources are required? What are recent improvements? What are the next steps and the long-term goals of the development?

This minisymposium brings together experts for different software in the context of UQ, ranging from tools that ease up individual tasks of UQ (such as surrogate modelling, UQ workflows, dimensionality reduction, data augmentation) up to whole frameworks for solving UQ problems. The minisymposium will foster discussion and exchange of ideas between developers and (prospective) users.

14:00

Quasi-Optimal Sparse Grids Method for Periodic Functions

14:30

Using Chaospy to address stochastically dependent Polynomial Chaos Expansions

15:00

UQEF: A software framework for UQ with automatic, scalable scheduling using Chaospy

15:30

Uncertainpy: A Python toolbox for uncertainty quantification and sensitivity analysis tailored towards computational neuroscience models

14:00

iCal
Harald Oberhauser | University of Oxford | United Kingdom

Alessandro Barp | University of Cambridge & Imperial College London | United Kingdom

George Wynne | Imperial College London | United Kingdom

Alex Lambert | Télécom Paris | France

MS341: Kernel Methods in Uncertainty Quantification Part I: Kernel-Based Distances, Hypothesis Tests and Statistical Estimators

Chair(s)
Francois-Xavier Briol (University College London and Turing Institute)

George Wynne (Imperial College London)

George Wynne (Imperial College London)

Room:
Exzellenzzentrum 0003

Topic:
Statistical methods for UQ

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Reproducing kernel Hilbert spaces are ubiquitous in applied mathematics and statistics due to the tractability provided by the reproducing property. They commonly underpin the theoretical analysis of stochastic processes (including Gaussian processes) and have historically been used to construct a variety of numerical schemes for interpolation, integration or solving differential equations. Additionally, within the machine learning literature, the last decade has seen fruitful research into the use of kernels in statistical tests, statistical estimators and sampling methods.

In this two-part mini-symposium, we propose to explore these more recent works and highlight their relevance to uncertainty quantification. The first session will focus on the use of kernel-based probability metrics and statistical divergences to construct statistical estimators and hypothesis tests for high-dimensional models or models with intractable likelihoods. The second session will focus on applications of kernels to problems in Monte Carlo methods and approximation of probability measures.

14:00

Learning Laws of Stochastic Processes

14:30

Statistical Inference for Generative Models with Maximum Mean Discrepancy

15:00

Kernel-based Statistical Tests on Infinite Dimensional Data

15:30

- NEW - Conditional quantile function estimation as an infinite task learning problem

16:30

iCal
Karina Koval | Courant Institute of Mathematical Sciences - New York University | United States

Joakim Beck | King Abdullah University of Science and Technology (KAUST) | Saudi Arabia

Julianne Chung | Virginia Polytechnic Institute and State University | United States

Quan Long | United Technology Research Center (UTRC) | United States

MS012: Recent advances and challenges in optimal experimental design for large-scale inverse problems (Part II of II)

Chair(s)
Ahmed Attia (Argonne National Laboratory)

Alen Alexanderian (North Carolina State University)

Alen Alexanderian (North Carolina State University)

Room:
MW HS 2001

Topic:
Design of experiments

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Computer models play an essential role in forecasting complicated phenomena such as the atmosphere, ocean dynamics, seismology among others. These models, however, are typically imperfect due to various sources of uncertainty. Measurements are snapshots of reality that are collected as an additional source of information and are used to update and even correct the model-based simulations or forecasts. The accuracy of the overall simulations and model-based forecasts is greatly influenced by the quality of the observational grid design used to collect measurements. Optimal data acquisition can be formulated as an optimal experimental design (OED) problem. The framework of model-based OED has gained wide popularity and attention from researchers in various fields in statistics, engineering, applied math and others. Challenges in model-based OED include high-dimensionality, misrepresentation of prior knowledge, increasing deviation from Gaussianity, high correlations of spatiotemporal observations, among others. This minisymposium aims to showcase the latest developments in tackling the challenges in the field of model-based OED for large-scale inverse problems.

16:30

- CANCELED - Optimal experimental design under model uncertainty, with application to subsurface flow

17:00

- CANCELED - Multilevel estimation of the expected information gain

17:30

- MOVED from MS011 - Enhanced Hybrid Projection Methods with Recycling for Large Inverse Problems

18:00

Optimal Bayesian Experimental Design Using Generalized Laplace Method

16:30

iCal
Stefano Pagani | Politecnico di Milano | Italy

Caterina Bigoni | Ecole polytechnique fédérale de Lausanne (EPFL) | Switzerland

Kathleen Champion | University of Washington | United States

Deep Ray | Rice University | United States

MS672: Achieving a data-model synergy in UQ (Part II of II)

Room:
MW HS 0001

Topic:
Reduced order models

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Recent years have seen the flourishing of techniques devoted to best incorporate data in the models, either for the solution of inverse problems or for approximation purposes. This includes domain-aware Machine Learning techniques, dynamic mode decomposition or data driven model order reduction methods. This minisymposium aims to provide a venue for young researchers focusing on the theoretical analysis, the development and the application of these methodologies.

16:30

Data-model integration in cardiac electrophysiology

17:00

Optimal Placement of Sensors to Assess Structural Damages

17:30

Data-driven discovery of coordinates for parsimonious dynamical models

18:00

Controlling oscillations in spectral schemes using deep learning

16:30

iCal
Jouni Susiluoto | Jet Propulsion Laboratory, California Institute of Technology | United States

Jenný Brynjarsdóttir | Case Western Reserve University | United States

Patrick Sheese | University of Toronto | Canada

Heikki Haario | Lappeenranta University of Technology | Finland

MS443: Uncertainty Quantification for Earth Remote Sensing (Part III of III)

Room:
MW HS 1801

Topic:
Inverse problems

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Numerous Earth-observing satellites provide high-resolution and high-volume data that facilitate scientific inference on physical and environmental processes. Most remote sensing data products used for scientific investigations are often subject to multiple stages of processing before they are widely used, and the scientific utility of these data products critically depends on a comprehensive assessment of the sources of uncertainty encountered in these stages of processing. One key stage involves the use of a retrieval algorithm to infer a geophysical quantity of interest from a satellite’s observed intensity of radiation.

The retrieval is an inverse problem that has been implemented mathematically and computationally in numerous ways for different satellite missions. Several of the presentations in this mini-symposium will each highlight an individual Earth-observing satellite and its retrieval methodology, emphasizing important contributions to uncertainty in retrieval data products. Methodological developments that interrogate the joint distribution of true geophysical states, retrieved states, and observed satellite spectra will be introduced. The presentations will span multiple Earth science applications, including weather and climate, the carbon cycle, air quality, atmospheric chemistry, and ecosystem health.

16:30

satGP: Efficient Gaussian process regression for massive remote sensing data

17:00

Model discrepancy in CO2 retrievals from the OCO-2 satellite

17:30

Uncertainty quantification of ACE-FTS data products

18:00

Emissions from individual power plants by satellite data

16:30

iCal
Bojana Rosic | University of Twente | Netherlands

Zaid Sawlan | KAUST | Saudi Arabia

Davide Baroli | Aachen Institute for Advanced Study in Computational Engineering Science | Germany

MS041: Propagation of uncertainties and parameter inference in material science (Part I of III)

Chair(s)
Alexander Litvinenko (RWTH Aachen)

Bojana Rosic (University of Twente)

Sebastian Krumscheid (RWTH Aachen)

Bojana Rosic (University of Twente)

Sebastian Krumscheid (RWTH Aachen)

Room:
MW HS 0350

Topic:
UQ for complex systems

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Organized in co-operation with

GAMM AG UQ

The MS focuses on the process of modeling, quantifying and estimating the effects of uncertainties that characterize irreversible/dissipative material behavior in quasi-static and dynamic conditions. Particular examples of great significance include metal fatigue and concrete fracture analysis, as well as material aging of bone tissues. Moreover, special attention will be paid to the multi-scale and multi-fidelity nature of these problems, as well as to Bayesian analysis and corresponding design of experiments. Numerical tools to be discussed are low-rank functional approximations, Bayesian learning, optimization, stochastic Galerkin, polynomial chaos expansion, and stochastic homogenization, to name just a few.

16:30

Stochastic multiscale analysis of nonlinear dissipative phenomena

17:30

Modelling fatigue crack initiation in metallic specimens by spatial Poisson processes

18:00

An uncertainty assessment framework for diffusive transport in biobased hydrogels

16:30

iCal
Michael Multerer | Università della Svizzera italiana | Switzerland

Martin Eigel | WIAS Berlin | Germany

Brendan Keith | TU München | Germany

MS463: Shape uncertainty quantification and applications (Part III of III)

Room:
MW HS 0250

Topic:
Uncertainty propagation

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Organized in co-operation with

GAMM AG UQ

Partial differential equations are a versatile tool to model and eventually simulate physical phenomena. An important aspect in view of the reliability and relevance of such simulations are uncertainties arising from unknown parameters and measurement errors. In particular, the modelling and discretization of uncertainties of the computational domain requires special care. Such uncertainties emerge in a natural fashion when considering products fabricated by line production which are subject to manufacturing tolerances or shapes which are obtained by remote sensing techniques, like e.g. ultrasound or magnetic resonance imaging. This minisymposium is dedicated to recent developments in the numerical treatment of shape uncertainties in partial differential equations and welcomes contributions addressing analytical aspects, forward modelling, assimilation of measurement data, optimization, and applications.

16:30

Numerical aspects of the domain mapping method for elliptic PDEs on random domains

17:00

Adaptive Stochastic Galerkin FEM for randomly perturbed domains

18:00

Risk Averse Design of Tall Buildings Under Uncertain Wind Loading

16:30

iCal
Julian Köllermeier | KU Leuven | Belgium

Giacomo Albi | University of Verona | Italy

Giacomo Dimarco | University of Ferrara | Italy

Yuhua Zhu | Stanford University | United States

MS553: Advances in uncertainty quantification for kinetic and transport phenomena (Part III of III)

Room:
MW HS 1250

Topic:
Multiscale UQ methods

Form of presentation:
Mini-symposium

Duration:
120 Minutes

In this session we concentrate on the latest research insights for uncertainty quantification in transport problems and high-dimensional systems under structural uncertainties, with focus on kinetic and hyperbolic PDEs and multiscale interacting particle systems.

16:30

- NEW - On new developments for Hyperbolic Moment Models of rarefied gases and the connection to Uncertainty Quantification

17:00

Mean-field feedback stabilization of collective behavior with uncertainty

17:30

Micro-macro generalized polynomial chaos techniques for kinetic equations

18:00

- CANCELED - A consensus-based global optimization method for high dimensional machine learning problems

16:30

iCal
Adrian Sandu | Virginia Tech | United States

Melina Freitag | University of Potsdam | Germany

Răzvan Ştefănescu | Spire Global, Inc. | United States

Manuel Pulido | Universidad Nacional del Nordeste | Argentina

MS092: Algorithms for Large Scale and Non-linear Data Assimilation (Part II of II)

Room:
MW ZS 1050

Topic:
Data assimilation

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Accuracy is always at odds with efficiency in the context of Data Assimilation on complex dynamical system. Such systems often involve large amounts of variables, with impactful non-linearities, and poorly understood stochastic behaviour. Tackling these problems in an efficient manner is the key to unlocking the next generation of algorithms. The discussion of directions such as exploiting the time-dependent structure of natural systems, reduced order modeling, accounting for model error, and efficient ways to solve the underlying optimization problem, are just some of the topics of fundamental importance in the next few years of research, that will be covered.

16:30

Data assimilation for constructing stabilized POD models

17:00

Solving weak constraint variational data assimilation problems using a low-rank approach

17:30

- CANCELED - Data Thinning Strategies for Global Assimilation of Large Number of Radio Occultation Profiles

18:00

Model error covariance estimation using the Expectation-Maximization algorithm in sequential Monte Carlo and ensemble Kalman filters

16:30

iCal
Jake Roth | Argonne National Laboratory | United States

Kenan Sehic | Technical University of Denmark | Denmark

Tobias Grafke | University of Warwick | United Kingdom

Charlotte Haley | Argonne National Laboratory | United States

MS871: Characterization and prediction of rare and extreme events in complex systems

Chair(s)
Vishwas Rao (Argonne National Laboratory)

Charlotte Haley (Argonne National Laboratory)

Mihai Anitescu (Argonne National Laboratory and University of Chicago)

Charlotte Haley (Argonne National Laboratory)

Mihai Anitescu (Argonne National Laboratory and University of Chicago)

Room:
MW ZS 2050

Topic:
Rare events and Risk

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Characterization and prediction of rare and extreme events that correspond to large excursions is of

central importance in several applications. Important examples can be found in natural phenomena

such as climate, weather, oceanography and engineering systems such as structures, power grids, etc.

Accurate characterization and reliable prediction of these events allows for realistic balancing of

risks and costs in complex and expensive infrastructure. Two important challenges related to these

rare and extreme events are i) limited availability of data that correspond to these rare and extreme

events, leads to difficulty in quantifying tail properties for the relevant distribution and ii)

determining the statistics of level crossings and durations of temporally or spatially correlated

processes. The aim of this MS is to present research that address these two general problems. Approaches based on, but not restricted to, data, dynamics, or a combination of both will be discussed.

16:30

- NEW - Characterizing the distribution of cascading power network failures

17:00

Estimation of Failure Probabilities via Local MCMC Subset Approximations

17:30

Extreme Event Quantification for Rogue Waves in Deep Sea

18:00

Extreme values of LIDAR wind speed

16:30

iCal
Arindam Fadikar | Argonne National Lab | United States

Matthias Poloczek | Uber | United States

Max Balandat | Facebook | United States

Mickael Binois | INRA Sophia Antipolis Mediterranean | France

MS762: Statistical Surrogate Modeling and Optimization for Stochastic Simulation (Part II of II)

Room:
MW ZS 1450

Topic:
Surrogate models

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Historically, design and analysis of computer experiments focused on deterministic solvers from the physical sciences via Gaussian process (GP) interpolation. But nowadays computer modeling is common in the social, management and biological sciences, where stochastic simulations abound. In this minisymposium, we bring together a selection of researchers in the areas of statistical surrogate modeling, active learning, and Bayesian optimization of stochastic computer model, simulation campaigns, and high volume observational studies. Noisier simulations demand bigger experiments to isolate signal from noise, and more sophisticated GP models -- such as adding a variance processes to track changes in noise throughout the input space in the face of heteroskedasticity. Appropriate surrogate modeling is key to the propagation of uncertainty to decision criteria underlying important large-scale and real time control of systems which rely on expensive simulation campaigns. Think of synthesis between off-line simulation of urban road traffic and ride demand with on-line measurements from potential riders and their routes in the assignment of a car. Or similarly the combination of limited data on disease spread combined with social-network backed simulation of epidemiological dynamics and entertainment of intervention strategies such as vaccination and quarantine. The talks will be on these methodologies and applied in those challenging modeling and optimization real-world problems.

16:30

Quantile-based Gaussian Process Emulation for a Stochastic Agent Based Model

17:00

Efficient optimization of high-dimensional expensive functions

17:30

A Differentiable Programming Approach to Bayesian Optimization

18:00

Bayesian Optimization and Dimension Reduction with Active Subspaces

16:30

iCal
Sayan Mukherjee | Duke University | United States

Stefan Klus | Freie Universitat Berlin | Germany

Ben Zhang | Massachusetts Institute of Technology | United States

Robert Webber | New York University | United States

MS122: Computational Statistics meets Computational Dynamics (Part II of II)

Chair(s)
Ben Zhang (Massachusetts Institute of Technology)

Tuhin Sahai (United Technologies Research Center)

Kevin Lin (University of Arizona)

Tuhin Sahai (United Technologies Research Center)

Kevin Lin (University of Arizona)

Room:
MW ZS 1550

Topic:
Statistical methods for UQ

Form of presentation:
Mini-symposium

Duration:
120 Minutes

In this minisymposium, we explore the symbiotic relationship between computational statistics and computational dynamics. The interaction of the two fields have long been established. Efficiently computing statistics of dynamical quantities is of interest in science and engineering, and cleverly constructed dynamical systems are used to sample from high-dimensional probability distributions. We will highlight recent advances in numerical methods that utilize tools in one field to solve problems in the other in a novel fashion. We will exhibit new algorithms for chaotic sensitivity analysis, rare event simulation, stochastic optimal control, and data assimilation.

16:30

Gibbs posterior convergence and the thermodynamic formalism

17:00

Data-driven approximation of the Koopman generator: Model reduction, system identification, and control

17:30

Efficient sampling methods for stochastic dynamical systems using Koopman eigenfunctions

18:00

The role of lag time in spectral estimation for Markov processes

16:30

iCal
Alexander Gilbert | Ruprecht-Karls-Universität Heidelberg | Germany

Peter Zaspel | Jacobs University | Germany

Chuntao Chen | Monash University | Australia

Michael Brennan | Massachusetts Institute of Technology | United States

MS812: Multilevel and Multi-fidelity Methods for Model-Based Statistical Learning (Part II of III)

Chair(s)
Tiangang Cui (Monash University)

Santiago Badia (Monash University)

Alexander Gilbert (Ruprecht-Karls University Heidelberg)

Youssef Marzouk (Massachusetts Institute of Technology)

Robert Scheichl (Ruprecht-Karls University Heidelberg)

Santiago Badia (Monash University)

Alexander Gilbert (Ruprecht-Karls University Heidelberg)

Youssef Marzouk (Massachusetts Institute of Technology)

Robert Scheichl (Ruprecht-Karls University Heidelberg)

Room:
MW 2250

Topic:
Statistical methods for UQ

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Organized in co-operation with

GAMM AG UQ

The synthesis of various information sources, including a priori domain knowledge, statistical assumptions, field data, etc., large-scale numerical models is one of the key steps in building interpretable and predictive models for supporting critical decisions in science, engineering, medicine, and beyond. Typical examples can be found in oil/gas reservoir modeling, treatment of saltwater intrusion, medical imaging, tumor treatment, aircraft design. Because of the computationally costly nature of the numerical models and stringent requirements on the accuracy of the statistical learning outcomes, multilevel and multi-fidelity methods provide a viable route for solving these model-based statistical learning tasks. This mini-symposium will bring together researchers working on the forefront of multilevel and multi-fidelity methods (and other relevant methods) intended to accelerate model-based statistical learning tasks.

16:30

Multilevel quasi-Monte Carlo methods for random elliptic eigenvalue problems

17:00

Multifidelity machine learning by the sparse grid combination technique

17:30

Multi-Level Optimization Based Monte-Carlo Samplers for Large-Scale Inverse Problems

18:00

Hierarchically Structured Transport Maps for Inference Problems

16:30

iCal
Chiheb Ben Hammouda | King Abdullah University of Science and Technology (KAUST) | Saudi Arabia

Aitor Muguruza Gonzalez | Imperial College London | United Kingdom

Jérôme Lelong | Université Grenoble Alpes | France

Blanka Horvath | King’s College London | United Kingdom

MS172: UQ for Rough Volatility and Predictive Models in Finance (Part II of II)

Room:
MW 1701

Topic:
UQ for complex systems

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Rough volatility models are an increasingly popular class of models in quantitative finance. In contrast to conventional stochastic volatility models, the volatility is driven by a fractional Brownian motion with Hurst index H < 1/2 which is rougher than Brownian motion. This change greatly improves the fit to time series data of underlying asset prices as well as to option prices, see, for instance, [Bayer, Friz, Gatheral, Quantitative Finance 16(6), 887-904, 2016]. Hence, introducing non-Markovian noise improves the predictive power of the model while maintaining parsimoniousness. Unfortunately, the loss of the Markov property poses severe challenges for theoretical and numerical analyses as well as for computational practice.

This minisymposium brings together different approaches for various UQ tasks in the context of rough volatility models and predictive models in finance. The problems addressed range from calibration and statistical analysis of the model parameters to optimal control of rough volatility models. To overcome the considerable practical hurdles posed by the lack of Markovianity, the contributions to the minisymposium use diverse tools such as deep neural networks and large deviation theory, assisted by properly analyzed simulation techniques.

16:30

- CANCELED - Hierarchical adaptive sparse grids and quasi Monte Carlo for option pricing under the rough Bergomi model

17:00

Rough Volatility: A Measure-Change Point of View

17:30

Pricing path-dependent Bermudan options: an embarrassingly parallel algorithm

18:00

On deep calibration of (rough) stochastic volatility models

16:30

iCal
Shunzhou Wan | University College London | United Kingdom

Jalal Lakhlili | Max Planck Institute for Plasma Physics | Germany

Wouter Edeling | Centrum Wiskunde & Informatica | Netherlands

MS651: Tools for enabling Verification, Validation and Uncertainty Quantification in multiscale simulations and workflows

Room:
MW HS 0337

Topic:
Multiscale UQ methods

Form of presentation:
Mini-symposium

Duration:
120 Minutes

[ Moved from MW HS 2235 ]

Data-science and numerical simulation are moving rapidly toward a workflow based approach for complex multiscale or multiphysics problems, which better suits the many-tasks paradigm followed by HPC centers on the path to exascale. As a result, a wide range of tools and frameworks (both generic and domain specific) have been developed over the years in order to support scientists in designing, implementing and running their complex simulations and workflows efficiently on HPC systems.

In order to produce "actionable" results, these simulations and workflows need to be validated, verified and equipped with uncertainty quantification (VVUQ) such that their output may be relied upon when making important decisions in various domains. The VECMA project (https://www.vecma.eu) aims at developing an open source toolkit (https://www.vecma-toolkit.eu) to ease, and automate where possible, the addition of VVUQ into such multiscale or multiphysics simulations.

In this minisymposium we invite developers of this toolkit to present its most recent version, and researchers in various domains (Fusion, Materials, Climate, Bio-medicine, etc...) to present how it can be integrated into existing applications in order to add VVUQ capabilities.

17:00

Verification, Validation & Uncertainty Quantification for Molecular Dynamics Simulation

17:30

VVUQ tools applied to fusion multiscale workflow simulations

18:00

Deriving reduced subgrid scale models from data

16:30

iCal
Jacob Sturdy | Norwegian University of Science and Technology | Norway

Huijuan Xu | Georgia Institute of Technology | United States

Mitchel Colebank | North Carolina State University | United States

Gabriel Maher | Stanford University | United States

MS582: Advances in Inverse Problems and Uncertainty Quantification in Cardiovascular Modeling (Part II of II)

Chair(s)
Mitchel Colebank (North Carolina State University)

Dirk Husmeier (School of Mathematics and Statistics, University of Glasgow)

Mette Olufsen (North Carolina State University)

Dirk Husmeier (School of Mathematics and Statistics, University of Glasgow)

Mette Olufsen (North Carolina State University)

Room:
IAS 0.001

Topic:
UQ for complex systems

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Advances in computational medicine have made mathematical modeling of hemodynamics a key area of scientific research. Innovations in high performance computing and high-fidelity models allows for sophisticated approximations of in-vivo cardiovascular dynamics. To this end, a variety of models including system level 0D models, 1D fluid dynamics network models, and 3D fluid structure interaction models, can be used to investigate structure-function relation of the cardiovascular system, on a local, global, or multiscale level. However, these computational models are susceptible to both model discrepancy and uncertainty in model inputs, and predictions. Cardiovascular models are calibrated to sparse data, i.e. they contain parameters unmeasurable in-vivo, making parameter estimation and forward uncertainty propagation difficult. This minisymposium will focus on cardiovascular inverse problems and statistical inference methodology including:

• Parameter estimation techniques for complex ODE-PDE coupled models

• Novel emulation and metamodeling procedures for high-fidelity models

• Advances in surrogate and low-fidelity model construction

• Quantification of model consistency using machine-learning

• Efficient uncertainty propagation and quantification

• Innovative numerical and analytical sensitivity techniques

16:30

Sensitivity analysis informed cardiovascular modelling for clinical applications

17:00

- CANCELED - Uncertainty Quantification in Numerical Simulations of the Hemodynamics of the Aorta

17:30

Parameter identifiability and estimation in a 1D cardiovascular fluid dynamics model

18:00

Anatomic Model Uncertainty through Convolutional Bayesian Dropout Networks

16:30

iCal
Aaron Smith | University of Ottawa | Canada

Viacheslav Natarovskii | Georg-August University of Göttingen | Germany

Alexandre Thiery | National University of Singapore | Singapore

Kengo Kamatani | Osaka University | Japan

MS141: Recent advances in Markov chain Monte Carlo

Chair(s)
Björn Sprungk (TU Bergakademie Freiberg)

Daniel Rudolf (Georg-August University of Göttingen)

Daniel Rudolf (Georg-August University of Göttingen)

Room:
IAS 4.001

Topic:
Inverse problems

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Organized in co-operation with

GAMM AG UQ

Markov chain Monte Carlo (MCMC) methods have become a well-established tool for approximate sampling in computational statistics and are indespensible for the quantification of uncertainty. One of the main advantages of MCMC approaches, shown in numerical experiments and proven theoretically, is the robust behavior w.r.t. the dimension, such that in high-dimensional scenarios those are often the method of choice. In the recent years there have been several new advances in algorithmic as well as theoretical aspects of these methods. For instance, Wasserstein contraction in order to prove ergodicity of Markov chains, MCMC methods for inference on non-Euclidean spaces such as Riemannian manifolds, and efficient Metropolis-Hastings algorithms for highly concentrated target measures.

The goal of this session is to discuss these recent developments and the dimension dependence of MCMC from a theoretical as well as a practical point of view.

16:30

Rapid mixing of geodesic walks on manifolds with positive curvature

17:00

Quantitative convergence properties of slice sampling

17:30

Exploiting geometry for walking larger steps in Bayesian Inverse Problems

18:00

Robust Markov chain Monte Carlo methods with respect to tail and scaling properties

16:30

iCal
Yibo Yang | University of Pennsylvania | United States

Marton Havasi | University of Cambridge | United Kingdom

Ludger Paehler | Technical University of Munich | Germany

Nicholas Zabaras | University of Notre Dame | United States

MS693: Uncertainty Quantification in Deep Learning (Part III of III)

Chair(s)
Paris Perdikaris (University of Pennsylvania)

Phaedon-Stelios Koutsourelakis (Technical University of Munich)

Phaedon-Stelios Koutsourelakis (Technical University of Munich)

Room:
Interims Lecture Hall 101

Topic:
Physics-informed and data-driven UQ methods

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Organized in co-operation with

GAMM AG UQ

Deep learning techniques are becoming the center of attention across many scientific disciplines. Many predictive tasks are currently being tackled using over-parameterized, black-box discriminative models such as deep neural networks, in which interpretability and robustness is often sacrificed in favor of flexibility in representation and scalability in computation. Such models have yielded remarkable results in data-rich domains, yet their effectiveness in data-scarce and risk-sensitive tasks still remains questionable, primarily due to open challenges in statistical inference and uncertainty quantification. This mini-symposium invites contributions on uncertainty quantification methods for deep learning and their application in the physical and engineering sciences. Topics include (but are not limited to) Bayesian neural networks, deep generative models, posterior inference techniques, and applications to forward/inverse problems, active learning, Bayesian optimization and reinforcement learning.

16:30

Sampling the posterior of Bayesian neural networks, with neural networks

17:00

Refining the variational posterior through iterative optimization

17:30

Uncertainties in Attention-based Koopman Embeddings

18:00

Integration of adversarial autoencoders with residual dense convolutional networks for estimation of non-Gaussian conductivities in solute transport modeling

16:30

iCal
Brian M. Adams | Sandia National Laboratories | United States

Dirk Pflüger | University of Stuttgart | Germany

Charles H. Tong | Lawrence Livermore National Laboratory | United States

Stefano Marelli | ETH Zurich | Switzerland

MS823: Software for UQ (Part III of III)

Chair(s)
Tobias Neckel (Technical University of Munich)

Dirk Pflüger (University of Stuttgart)

Stefano Marelli (ETH Zurich)

Edoardo Patelli (Strathclyde University)

Dirk Pflüger (University of Stuttgart)

Stefano Marelli (ETH Zurich)

Edoardo Patelli (Strathclyde University)

Room:
Interims Lecture Hall 102

Topic:
Other

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Organized in co-operation with

GAMM AG UQ

With the ever increasing importance of UQ in various disciplines and fields, software solutions and libraries for UQ problems get more and more important. Progress and use of UQ techniques relies on the availability of software features and support. This raises interesting questions for the UQ community such as: What are the current properties of available tools? For which classes of problems have they been developed? What methods or algorithms do they provide? What are challenges for UQ software and which resources are required? What are recent improvements? What are the next steps and the long-term goals of the development?

This minisymposium brings together experts for different software in the context of UQ, ranging from tools that ease up individual tasks of UQ (such as surrogate modelling, UQ workflows, dimensionality reduction, data augmentation) up to whole frameworks for solving UQ problems. The minisymposium will foster discussion and exchange of ideas between developers and (prospective) users.

16:30

- CANCELED - New Ways to Explore and Predict with Dakota

17:00

Data-Driven Uncertainty Quantification with SG++

17:30

Recent Development in the PSUADE UQ Software

18:00

Blurring the lines between UQ and ML: a software perspective

16:30

iCal
Rémi Bardenet | CNRS & CRIStAL, Universite de Lille | France

Leah South | Lancaster University | United Kingdom

Nikolas Nüsken | Universität Potsdam | Germany

Carl-Johann Simon-Gabriel | ETH Zurich | Switzerland

MS342: Kernel Methods in Uncertainty Quantification Part II: Monte Carlo Methods and Approximation of Probability Measures

Chair(s)
Francois-Xavier Briol (University College London and Turing Institute)

George Wynne (Imperial College London)

George Wynne (Imperial College London)

Room:
Exzellenzzentrum 0003

Topic:
Statistical methods for UQ

Form of presentation:
Mini-symposium

Duration:
120 Minutes

Reproducing kernel Hilbert spaces are ubiquitous in applied mathematics and statistics due to the tractability provided by the reproducing property. They commonly underpin the theoretical analysis of stochastic processes (including Gaussian processes) and have historically been used to construct a variety of numerical schemes for interpolation, integration or solving differential equations. Additionally, within the machine learning literature, the last decade has seen fruitful research into the use of kernels in statistical tests, statistical estimators and sampling methods.

In this two-part mini-symposium, we propose to explore these more recent works and highlight their relevance to uncertainty quantification. The first session will focus on the use of kernel-based probability metrics and statistical divergences to construct statistical estimators and hypothesis tests for high-dimensional models or models with intractable likelihoods. The second session will focus on applications of kernels to problems in Monte Carlo methods and approximation of probability measures.

16:30

DPPs: The Kernel Machine of Point Processes

17:00

Developments in Stein-Based Control Variates

17:30

On the Geometry of Stein Variational Gradient Descent

18:00

- CANCELED - Kernel Distribution Embeddings: Universal Kernels, Characteristic Kernels and Kernel Metrics on Distributions