Parts | Days | Selection | Search | Updates | Downloads | Help
O: Fachverband Oberflächenphysik
O 71: Mini-Symposium: Machine learning applications in surface science I
O 71.5: Invited Talk
Wednesday, March 3, 2021, 15:00–15:30, R1
Theory-informed Machine Learning for Surface and Interface Structure Reconstruction from Experimental Data — Eric Schwenker1,2, Chaitanya Kolluru1,3, Marcel Chlupsa1, Arun Mannodi Kanakkithodi1, Richard Hennig3, Pierre Darancet1,2, and •Maria Chan1,2 — 1Argonne National Laboratory, Lemont, USA — 2Northwestern University, Evanston, USA — 3University of Florida, Gainsville, USA
Determining atomistic structure at surfaces and interfaces is challenging because metastable surfaces/interfaces are likely accessible under realistic conditions, rendering energy-only searches insufficient, and experimental data often give incomplete information. Therefore, neither theory nor experimental data alone is sufficient to determine these structures. In this talk, we will discuss how we use machine learning to combine experimental and theory-based data to determine surface and interface structures.