Join us for a seminar with Professor Taylor D. Sparks, Associate Professor in Materials Science and Engineering at the University of Utah. The seminar will focus on the impact of machine learning on crystal structure prediction.
ABSTRACT:
Crystal structure prediction has long fascinated scientists. There has been intense investigation over the last century ranging from simplistic rules to data-driven predictions and, most recently, generative artificial intelligence tools developed by academics and now deployed at scale by private companies like DeepMind. In this talk, Professor Sparks will describe the timeline of crystal structure prediction and describe how machine learning has supplemented and, in some cases, replaced traditional approaches. Professor Sparks will compare generative models including variational autoencoders, generative adversarial networks, and diffusion models and describe new efforts to condition these models to achieve inverse design of new crystal structures. Professor Sparks will give specific examples of their xtal2png and CrysTens representations and their machine learning contributions to greatly accelerate the Flexible Unit Structure Engine (FUSE) software package.