BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Atomistic Machine Learning between Physics and Data - Prof. Michel
 e Ceriotti\, Head COSMO Laboratory
DTSTART:20190424T100000Z
DTEND:20190424T110000Z
UID:TALK111646@talks.cam.ac.uk
CONTACT:Nichola Daily
DESCRIPTION:Statistical regression techniques have become very fashionable
  as a tool to predict the properties of systems at the atomic scale\, side
 stepping much of the computational cost of first-principles simulations an
 d making it possible to perform simulations that require thorough statisti
 cal sampling without compromising on the accuracy of the electronic struct
 ure model.\n\nIn this talk I will argue how data-driven modelling can be r
 ooted in a mathematically rigorous and physically-motivated framework\, an
 d how this is beneficial to the accuracy and the transferability of the mo
 del. I will also highlight how machine learning - despite amounting essent
 ially to data interpolation - can provide important physical insights on t
 he behaviour of complex systems\, on the synthesizability and on the struc
 ture-property relations of materials.\n\nI will give examples concerning a
 ll sorts of atomistic systems\, from semiconductors to molecular crystals\
 , and properties as diverse as drug-protein interactions\, dielectric resp
 onse of aqueous systems and NMR chemical shielding in the solid state.\n
LOCATION:Goldsmiths 1\, Lecture Theatre\, Department of Materials Science 
 & Metallurgy
END:VEVENT
END:VCALENDAR