BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20220812T074334Z
LOCATION:Foyer 2nd Floor
DTSTART;TZID=Europe/Stockholm:20220628T090000
DTEND;TZID=Europe/Stockholm:20220628T110000
UID:submissions.pasc-conference.org_PASC22_sess181_pos122@linklings.com
SUMMARY:P11 - Real-Time Large Deformation Simulations using Probabilistic 
 Deep Learning Framework
DESCRIPTION:Poster\n\nP11 - Real-Time Large Deformation Simulations using 
 Probabilistic Deep Learning Framework\n\nDeshpande, Lengiewicz, Bordas\n\n
 Several engineering applications rely on the predictive capabilities of co
 mputational models. Some of these applications, like biomedical simulation
 s, require computationally efficient or even real-time solutions. Conventi
 onal methods for solving the underlying nonlinear problems, such as the Fi
 nite Element Method are computationally far too expensive. In this work, w
 e propose a probabilistic deep learning surrogate framework that is capabl
 e of accurately and efficiently predicting non-linear deformations of bodi
 es together with the predictions’ uncertainties. The framework uses a spec
 ial convolutional neural network architecture (U-Net), which has strong re
 semblances to Finite Element multigrid methods and proves to be capable of
  capturing non-linear responses characteristic to large deformation regime
 s. Our surrogate framework directly takes the Finite Element nodal forces 
 at the neural network input to give nodal displacements at its output. The
  probabilistic part of the framework is based on a dedicated Variational I
 nference formulation, with which we are not only able to efficiently captu
 re uncertainties related to noisy data, but we also have knowledge about t
 he model uncertainties—which is especially important in regions not well s
 upported by the data (e.g., the extrapolated region). Hence our framework 
 acts as an important step towards making real-time large deformation simul
 ations more trustworthy.
END:VEVENT
END:VCALENDAR