BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:‘Machine-learning in chemistry from the bottom up’ - Professor
  Michele Ceriotti\, Institute of Materials\, EPFL
DTSTART:20230321T160000Z
DTEND:20230321T170000Z
UID:TALK198070@talks.cam.ac.uk
CONTACT:Chloe Barker
DESCRIPTION:Machine-learning techniques are often applied to perform "end-
 to-end" predictions\, that is to make a black-box estimate of a property o
 f interest using only a coarse description of the corresponding inputs.\nI
 n contrast\, atomic-scale modeling of matter is most useful when it allows
  to gather a mechanistic insight into the microscopic processes that under
 lie the behavior of molecules and materials. \nIn this talk I will provide
  an overview of the progress that has been made combining these two philos
 ophies\, using data-driven techniques to build surrogate models of the qua
 ntum mechanical behavior of atoms\, enabling "bottom-up" simulations that 
 reveal the behavior of matter in realistic conditions with uncompromising 
 accuracy. \nI will show that data-driven modeling can be rooted in a mathe
 matically rigorous and physically-motivated symmetry-adapted framework\, a
 nd discuss the benefits of taking a well-principled approach. \nI will pre
 sent several examples demonstrating how the combination of machine-learnin
 g and atomistic simulations can offer useful insights on the behavior of c
 omplex systems\, and discuss the challenges towards an integrated modeling
  framework in which physics-driven and data-driven steps can be combined t
 o improve the accuracy\, the computational efficiency and the transferabil
 ity of predictions\, from interatomic potentials to electronic-structure p
 roperties.
LOCATION:Dept of Chemistry\, Wolfson Lecture Theatre 
END:VEVENT
END:VCALENDAR