BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Projected variational inference for high-dimensional Bayesian inve
 rse problems - Peng Chen (Georgia Institute of Technology)
DTSTART:20230428T090000Z
DTEND:20230428T100000Z
UID:TALK198451@talks.cam.ac.uk
DESCRIPTION:In this talk\, I will present a class of transport-based proje
 cted variational inference methods to tackle the computational challenges 
 of the curse of dimensionality and unaffordable evaluation cost for high-d
 imensional Bayesian inverse problems governed by complex models. We projec
 t the high-dimensional parameters to intrinsically low-dimensional data-in
 formed subspaces and employ transport-based variational methods (Stein and
  Wasserstein variational gradient descent using kernels and neural network
 s) to push samples drawn from the prior to a projected posterior. Moreover
 \, we employ fast surrogate models to approximate the parameter-to-observa
 ble map. I will present error bounds for the projected posterior distribut
 ion measured in Kullback--Leibler divergence. Numerical experiments will b
 e presented to demonstrate the properties of our methods\, including impro
 ved accuracy\, fast convergence with complexity independent of the paramet
 er dimension and the number of samples\, strong parallel scalability in pr
 ocessor cores\, and weak data scalability in the data dimension.
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR