Computational materials engineering with active learning
17th. Nov. 2022, ICAMS-IM Seminar, Online Zoom links will be communicated
Start: 17th. Nov. 2022. 12:00
End: 17th. Nov. 2022. 01:00 p.m.
Milica Todorovic University of Turku, Finland
M. Stricker ICAMS, Ruhr-Universität Bochum
Data-driven materials science based on artificial intelligence (AI) algorithms has facilitated breakthroughs in materials optimization and design. Of particular interest are active learning algorithms, where datasets are collected by smart sampling on-the-fly in the search for optimal solutions. We encoded such a probabilistic algorithm into the Bayesian Optimization Structure Search (BOSS) Python tool for materials research. We utilized this versatile tool in computational studies of functional materials, like molecular surface adsorbates, thin films, solid-solid interfaces, molecular conformers, and even to optimise experimental outcomes. Agreement between optimal solutions and experimental measurements suggests that active learning is capable of good accuracy at computational costs up to 10 times smaller than other approaches. In design-of-experiment tasks, BO delivers predictive models to optimize materials, processes and devices, while conducting as few experiments as possible.
The Materials Science and Technology Seminar is jointly organized by the IM (Institute for Materials) and ICAMS (Interdisciplinary Centre for Advanced Materials Simulation). Members of the RUB Materials Research Department MRD and of the DGM Regionalforum Rhein-Ruhr are cordially invited to participate in the seminar. For further information please contact: Dr. Inmaculada López Galilea, email@example.com, phone: +49 234 32 25957.