BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Scientific Uses of Automatic Differentiation - Michael Brenner\, H
 arvard U.
DTSTART:20230217T160000Z
DTEND:20230217T170000Z
UID:TALK195031@talks.cam.ac.uk
CONTACT:Prof. Jerome Neufeld
DESCRIPTION:There is much excitement (some of it legitimate) about applica
 tions of machine learning to the sciences. Here I'm going to argue that a 
 primary opportunity is not machine learning per se\, but instead that the 
 tools underlying the ML revolution yield significant opportunities for sci
 entific  discovery. Primary among these tools is automatic differentiation
  and the scalability of codes.  Neural network architectures are similar t
 o time rollouts in dynamical systems\, and therefore the technical advance
 s underlying the ML have the potential to directly translate into  the abi
 lity to solve important optimization problems in the sciences that have he
 retofore not been tackled. I will describe a number of different direction
 s we have been undertaking using automatic differentiation and large scale
  optimization to solve science problems\, including developing new algorit
 hms for solving partial differential equations\, the design of energy land
 scapes and kinetic pathways for self assembly\, the design of fluids with 
 designer rheologies\, "optimal porous media"\, and learning the division r
 ules for models of tissue development.
LOCATION:MR2\, Centre for Mathematical Sciences\, Wilberforce Road\, Cambr
 idge
END:VEVENT
END:VCALENDAR