BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:R-adaptivity\, deep learning and the deep ritz method - Chris Budd
  (University of Bath)
DTSTART:20231019T140000Z
DTEND:20231019T150000Z
UID:TALK207412@talks.cam.ac.uk
CONTACT:Nicolas Boulle
DESCRIPTION:PINNS (physics informed neural nets) are becoming increasingly
  popular methods for using deep learning techniques to solve a wide variet
 y of differential equations. They have been advertised as 'mesh free metho
 ds' which can out perform traditional methods. But how good are they in pr
 actice? In this talk I will look at how they compare with traditional tech
 niques such as the finite element method on different types of PDE\, linki
 ng their performance to that of general  nonlinear approximation methods s
 uch as Free Knot Splines. I will show that a combination of 'traditional' 
 numerical analysis and  deep learning can yield good results. But there is
  still a lot to be learned about the performance and reliability of a PINN
  based method.\n\nJoint work with Simone Appela\, Teo Deveney\, and Lisa K
 reusser.
LOCATION:Centre for Mathematical Sciences\, MR14
END:VEVENT
END:VCALENDAR