BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Global optimisation of atomistic structure with evolutionary algor
 ithms and reinforcement learning - Bjørk Hammer\, Aarhus University
DTSTART:20210125T163000Z
DTEND:20210125T173000Z
UID:TALK155839@talks.cam.ac.uk
CONTACT:Bingqing Cheng
DESCRIPTION: Atomistic simulations of the physico-chemical processes at in
 organic surfaces often require knowledge of the energetically optimal stat
 e of the surfaces. In this talk\, examples are given of intricate surface 
 reconstructions and surprising shapes assumed by metal nano-particles supp
 orted on oxide surfaces. The focus of the talk will be on how to identify 
 such optimal structure given a costly total energy method\, typically base
 d on density functional theory (DFT). A number of approaches will be prese
 nted. 1) A purely evolutionary approach in which new structural candidates
  are created by random cross-over and mutation operations\, 2) a machine l
 earned-enhanced evolutionary approach in which an on-the-fly learned surro
 gate energy landscape directs the candidate production\, and finally 3) a 
 pure reinforcement learning approach in which image recognition via a conv
 olutionary neural network is used to build up rational knowledge about the
  energy landscape\, that eventually leads to construction of the globally 
 optimal structure.\n\n[1] Evolutionary approach (EA): Phys. Rev. Lett. 108
 \, 126101 (2012)\n[2] ML assisted EA: Phys. Rev. Lett. 124\, 086102 (2020)
  and https://gofee.au.dk\n[3] Reinforcement learning:  Phys. Rev. B\, 102\
 , 075427 (2020).
LOCATION:virtual ZOOM meeting ID: 263 591 6003\, Passcode: 000042\, https:
 //us02web.zoom.us/j/2635916003?pwd=ZlBEQnRENGwxNmJGMENGMWxjak5nUT09
END:VEVENT
END:VCALENDAR