BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Tutorial on Bayesian inference with deep generative models: Exampl
 es from hydrogeology and geophysics - Stig Niklas Linde (Université de La
 usanne)
DTSTART:20260211T143000Z
DTEND:20260211T153000Z
UID:TALK242332@talks.cam.ac.uk
DESCRIPTION:By measuring responses of physical systems\, it is possible to
  gain knowledge about their properties by inversion or their states by dat
 a assimilation. Many natural systems display strong spatial heterogeneity 
 and temporal dynamics acting over many scales. The information content in 
 the available noise-contaminated data is generally insufficient to derive 
 unique estimates at relevant spatial and temporal scales. Deep generative 
 modeling excels at generating (time-dependent) unconditional realizations 
 of properties or states featuring similar characteristics as the training 
 data. This provides a new way to address Bayesian inverse problems (and da
 ta assimilation) in which a deep generative model representing prior knowl
 edge is combined with a likelihood term accounting for the available site-
 specific data to estimate the posterior distribution. Using examples from 
 hydrogeology and geophysics\, I will review and discuss some of the more c
 ommon deep generative models (variational autoencoders\, generative advers
 arial networks\, diffusion models) to represent prior knowledge\, as well 
 as the strengths and weaknesses of alternative approaches to estimate the 
 posterior distribution (Markov chain Monte Carlo\, sequential Monte Carlo\
 , variational Bayes\, diffusion posterior sampling). I will end by highlig
 hting the main challenges of such inversion workflows.
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR