Mathematical Foundations for Data-driven Engineering

Home > Mathematical Foundations for Data-driven Engineering

Mathematical Foundations for Data-driven Engineering

 09 - 12 May 2023

ICMS, Bayes Centre, Edinburgh

Scientific organisers

  • Jonas Latz, Heriot-Watt University
  • Eric Moulines, Ecole Polytechnique
  • Andrew Stuart , California Institute of Technology
  • Aretha Teckentrup , University of Edinburgh


This workshop was part of the Isaac Newton Institute programme on The mathematical and statistical foundation of future data-driven engineering:

In tandem with the workshop on Computational Challenges and Emerging Tools, this week-long workshop focused on algorithms and methodology to address key tasks in data-driven engineering, including:

  1. data assimilation and statistical inverse problems, 
  2. reinforcement learning and control, 
  3. model order reduction,
  4. bridging mechanistic and data-driven models and methods.

A particular focus of this workshop was on the underlying mathematical and statistical foundations required to make computational tools efficient and transferable.



Tuesday 9 May 2023
Registration and refreshments
Welcome and housekeeping (organisers and ICMS)
Susana Gomes, University of Warwick Parameter Estimation for Macroscopic Pedestrian Dynamics Models using Microscopic Data
Sebastian Reich, University of Potsdam Minimum variance estimation for continuous time data assimilation
Lunch break
Michela Ottobre, Heriot-Watt University On Multiscale dynamics
Dan Crisan, Imperial College London Data assimilation for a Quasi-Geostrophic Model with Circulation-Preserving Stochastic Transport Noise
Coffee break
Mateusz Majka, Heriot-Watt University Solving mean-field games via fictitious play and birth-death
Hanne Kekkonen, Delft University of Technology Public lecture: Mathematics of Images
Drinks reception at ICMS
Wednesday 10 May 2023
Richard Nickl, University of Cambridge On the computational complexity of MCMC in non-linear high-dimensional regression models
Margaret Trautner, Caltech Operator Learning for Multiscale PDEs in Solid Mechanics
Coffee break
Daniel Walter, Humboldt-Universität zu Berlin A closed loop learning approach for optimal feedback laws in nonlinear control problems
Group photo and lunch
Workshop dinner
Thursday 11 May 2023
Alexandros Beskos, UCL Manifold Markov chain Monte Carlo methods for Bayesian inference in diffusion models
Kostas Zygalakis , University of Edinburgh On the connections between sampling, optimization and (stochastic) differential equations
Coffee break
Sinho Chewi, Massachusetts Institute of Technology Faster high-accuracy log-concave sampling via algorithmic warm starts
Lunch break
Björn Sprungk, TU Bergakademie Freiberg Noise-level robust Markov chain Monte Carlo and pushforward Markov kernels
François-Xavier Briol, UCL Multilevel Bayesian quadrature
Coffee break
Jana de Wiljes, University of Potsdam Sequential Bayesian Learning
Poster session and drinks reception
Friday 12 May 2023
Tim Sullivan, University of Warwick Recent advances in the definition and stability of non-parametric MAP estimators
Robert Scheichl, Heidelberg University Surrogates Based on Low-Rank Tensor Approximation for Future Data-Driven Engineering
Coffee break
Karen Veroy-Grepl, Eindhoven University of Technology Model Order Reduction in Data Assimilation
Closing remarks
Take away lunch and end of workshop

Sponsors and Funders:

  • INI
  • GMJT
  • EMS
  • The