BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Moving beyond screening via generative machine learning models - P
 rof Taylor D. Sparks\, Materials Science &amp\; Engineering\, University o
 f Utah
DTSTART:20230206T143000Z
DTEND:20230206T150000Z
UID:TALK196666@talks.cam.ac.uk
CONTACT:Dr Christoph Schran
DESCRIPTION:Machine learning already enables the discovery of new material
 s by providing rapid predictions of properties to complement slower calcul
 ations and experiments. However\, a persistent criticism of machine learni
 ng enabled materials discovery is that new materials are very similar\, bo
 th chemically and structurally\, to previously known materials. This begs 
 the question “Can machine learning ever learn new chemistries and famili
 es of materials that differ from those present in the training data?” In
  this talk\, I will describe two important tools we are developing to trul
 y move beyond screening to actual discovery. First\, I will describe new g
 enerative machine learning approaches that can be used to generate structu
 res that do not yet exist\, but are likely to. I will compare generative m
 odels including variational autoencoders\, generative adversarial networks
 \, and diffusion models which have become standard in machine learning for
  images. I will describe the unique challenges that we face when using too
 ls of this nature to generate periodic crystalline structures. Second\, I 
 will describe the Descending from Stochastic Clustering Variance Regressio
 n (DiSCoVeR) algorithm to bias the discovery of new suggested materials aw
 ay from known chemistries in a systematic way towards unintuitive and even
  unlikely yet promising candidates for new materials.
LOCATION:Zoom link: https://zoom.us/j/92447982065?pwd=RkhaYkM5VTZPZ3pYSHpt
 UXlRSkppQT09
END:VEVENT
END:VCALENDAR