BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:PCSL x AI Center Seminar - AI for Science series - Dr. Oliver Dick
 s
DTSTART:20250408T160000
DTEND:20250408T170000
DTSTAMP:20260501T045121Z
UID:000f00defeca4331af6f3a0dc486567f83b5bd8812f732d31ae2fbbd
CATEGORIES:Conferences - Seminars
DESCRIPTION:Dr. Oliver Dicks\nThe talk is jointly organized by the PCSL L
 ab and the EPFL AI Center.\n\nFor on-site logistics\, please use the f
 ollowing form to register (with your EPFL email address): Here. \nThe t
 alk will be followed by a coffee session.\n\nHost: Dr. Daniel Korchinski\n
 \nTitle\nFinetuning Machine Learning Potentials for High Entropy Oxide Mat
 erial Discovery\n\nAbstract\nMACE is an equivariant message passing neural
  network (MPNN) machine learning interatomic potential (MLIP) that achieve
 s near quantum method calculation level accuracy for materials incorporati
 ng nearly the whole periodic table of elements for a fraction of the compu
 tational cost. This makes it ideal for exploring the huge compositional sp
 ace of the novel class of materials known as high entropy oxides (HEOs). H
 owever\, despite being trained on a database of over 1 million material sy
 stems whose properties have been calculated using density functional theor
 y\, there are still regions of chemical and structural space where its acc
 uracy is insufficient to estimate chemical stability. We demonstrate metho
 ds to construct minimal training databases to finetune MACE for specific t
 arget systems and elements to aid in high-throughput structure search for 
 brand new synthesizable materials not achievable with traditional methods\
 , and then subsequently calculate their properties with a high degree of a
 ccuracy.\n\nBio\nOliver Dicks is a computational physicist specializing in
  condensed matter theory and material discovery. He holds a Marie Sklodows
 ka-Curie (UKRI Guarantee) postdoctoral research fellowship at the Universi
 ty of British Columbia’s Quantum Matter Institute researching the use of
  machine learning potentials to model a new class of materials known as hi
 gh entropy oxides. He has worked to develop methods to calculate supercond
 ucting properties in high-throughput structure search applications\, in co
 llaboration with Intellectual Ventures. His additional scientific interest
 s include the effects of radiation damage on glass storage mediums for nuc
 lear waste\, as well as liquid matter theory and density functional theory
 . When not doing physics\, he enjoys running\, climbing\, and skiing throu
 gh the mountains of Canada. \n 
LOCATION:ELE 117 https://plan.epfl.ch/?room==ELE%20117 https://epfl.zoom.u
 s/j/65942852644?pwd=yqwO8MbnDAw59TwqCeZ0gMKEaA2cli.1
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR