BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Cambridge ELLIS seminar series – Dr Tian Xie – 23 Jan 2024 –
  2pm - Dr Tian Xie 
DTSTART:20240123T140000Z
DTEND:20240123T150000Z
UID:TALK209986@talks.cam.ac.uk
DESCRIPTION:The Cambridge ELLIS Unit Seminar Series holds talks by leading
  researchers in the area of machine learning and AI. Our next  speaker for
  2024 will be Dr. Tian Xie. Details of his talk can be found below. \n\nT
 itle: “MatterGen: a generative model for inorganic materials design"\n\n
 Abstract: The design of functional materials with desired properties is es
 sential in driving technological advances in areas like energy storage\, c
 atalysis\, and carbon capture. Traditionally\, materials design is achieve
 d by screening a large database of known materials and filtering down cand
 idates based on the application. Generative models provide a new paradigm 
 for materials design by directly generating entirely novel materials given
  desired property constraints. In this talk\, we present MatterGen\, a gen
 erative model that generates stable\, diverse inorganic materials across t
 he periodic table and can further be fine-tuned to steer the generation to
 wards a broad range of property constraints. To enable this\, we introduce
  a new diffusion-based generative process that produces crystalline struct
 ures by gradually refining atom types\, coordinates\, and the periodic lat
 tice. We further introduce adapter modules to enable fine-tuning towards a
 ny given property constraints with a labeled dataset. Compared to prior ge
 nerative models\, structures produced by MatterGen are more than twice as 
 likely to be novel and stable\, and more than 15 times closer to the local
  energy minimum. After fine-tuning\, MatterGen successfully generates stab
 le\, novel materials with desired chemistry\, symmetry\, as well as mechan
 ical\, electronic and magnetic properties. Finally\, we demonstrate multi-
 property materials design capabilities by proposing structures that have b
 oth high magnetic density and a chemical composition with low supply-chain
  risk. We believe that the quality of generated materials and the breadth 
 of MatterGen's capabilities represent a major advancement towards creating
  a universal generative model for materials design.\nhttps://eng-cam.zoom.
 us/j/81198870418?pwd=OFhHenUvM1JtWFlHRWt3aU12VkYxQT09 \n
LOCATION:https://eng-cam.zoom.us/j/81198870418?pwd=OFhHenUvM1JtWFlHRWt3aU1
 2VkYxQT09
END:VEVENT
END:VCALENDAR