BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Rothschild Lecture:  From Quantum Entanglement to Future Data-Driv
 en Engineering - Robert Scheichl (Ruprecht-Karls-Universität Heidelberg)
DTSTART:20230503T150000Z
DTEND:20230503T160000Z
UID:TALK199573@talks.cam.ac.uk
DESCRIPTION:In the last decade\, parallel to the rise of data science and 
 machine learning there has also been a vast growth in the interest and con
 tributions from numerical analysis and scientific computing in high-dimens
 ional Bayesian statistics\, in order to efficiently combine data and&nbsp\
 ;physical models to better understand and control engineering problems wit
 h a quantitative measure of the remaining uncertainty. Simply opening the 
 leading journals in the field or looking at recent job adverts will reveal
  this fact. But what are the problems and challenges that people are aimin
 g to address\, what are potential contributions and how can it&nbsp\;benef
 it the&nbsp\;field most effectively? In this talk\, I will try to summaris
 e some of the main areas of research where there are opportunities for num
 erical analysis to have an impact\, but also the difficulties and barriers
  encountered. More specifically\, I will present two exemplary approaches 
 that use surrogates to significantly accelerate Bayesian computation in hi
 gh-dimensional PDE-constrained applications: multilevel delayed acceptance
  MCMC [Lykkegaard et al\, 2023]\, as well as a measure-transport approach 
 based on low-rank tensor approximations [Cui et al\, 2022].
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR