BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Incorporating prior information into global search for improved st
 ructure prediction - Saki Shinoda (University of Cambridge)
DTSTART:20160309T120000Z
DTEND:20160309T121500Z
UID:TALK64966@talks.cam.ac.uk
CONTACT:Joseph Nelson
DESCRIPTION:General consensus holds that the more empirical information ne
 eded by a structure prediction method\, the less powerful it is\, in the s
 ense that it cannot predict a variety of structures from knowledge of chem
 ical composition alone.  However\, for almost all global search techniques
 \, incorporation of prior knowledge of some form improves performance by i
 ncreasing speed\, or where it can be evaluated\, improving accuracy.  Cons
 idering the computational complexity of the problem\, I argue that to move
  towards being able to carry out routine predictions of the possible struc
 tures of a system\, methods must be able to systematically take into accou
 nt what prior information already exists for the system.  By comparing dif
 ferent methods of energy landscape search methods in a common framework of
  general steps and considering (1) where and how different methods incorpo
 rate `prior knowledge'\, and (2) how these inclusions affect predictive po
 wer\, speed\, and accuracy\, we can look for explanations of existing stre
 ngths or means of potential improvement. This review focuses primarily on 
 ab initio applications of simulated annealing\, Monte Carlo basin-hopping\
 , evolutionary algorithms\, particle swarm optimisation\, and random searc
 h. 
LOCATION:TCM Seminar Room\, Cavendish Laboratory
END:VEVENT
END:VCALENDAR