Physics Colloquium: Applications of Artificial Intelligence to Neutron Scattering

Dr. William Ratcliff, NIST

Abstract:

Neutron scattering is a versatile technique for studying the structure and dynamics of materials. Unfortunately, there are a limited number of neutron sources available in the world to perform scientific experiments. In this talk, I will discuss the use of artificial intelligence to more efficiently use these instruments. Methods include reinforcement learning, demonstrated by Deep Mind in their alpha Go program, as well as techniques such as Bayesian optimization. I will also discuss how AI can accelerate the analysis of data that we take and aid our understanding.

Speaker bio:

Dr. William Ratcliff II was born in Ann Arbor, Michigan. He attended the University of Michigan for his undergraduate degree. After obtaining his Bachelors in Science and Engineering, he attended graduate school at Rutgers, working in the group of Professor Sang Wook Cheong on CMR manganites, dilute magnetic semiconductors, and frustrated magnets. After finishing his doctorate, he went to the NIST Center for Neutron Research in Gaithersburg, Maryland for a National Research Council Postdoc working with Dr. Seunghun Lee on frustrated magnets and multiferroic materials. Afterward, he joined the staff and has been at NIST for 19 years. Dr. Ratcliff has coauthored over 80 papers whose total citations exceed 5000, given several invited talks at international conferences, and organized workshops on magnetic structure determination. He is a two-time winner of the NIST Bronze Medal, the highest honorary recognition given by the institute. He is a fellow of the American Physical Society and an Associate Editor for Science Advances. He is past chair of the APS topical group on data science and is vice chair of the APS topical group on magnetism. His current research focuses on topological materials, multiferroic materials, and applications of AI to neutron scattering.