Abstract
Today’s ubiquitous data-driven workflows allow scientists to expand the limits of length and time scales in simulations. In my own field, namely first-principles atomistic simulations, the data itself is generated by systematic high-throughput workflows, which occupy a noteworthy chunk of the world’s supercomputing resources. Questions related to the efficiency, robustness and accuracy of simulation protocols and the reproducibility of obtained simulation data are thus more pressing than ever. Due to the complexity of underlying physical models (non-linear PDEs, multi-linear algebra, …) tackling these issues becomes inherently an interdisciplinary endeavour. After a brief introduction to the challenges of the field I will highlight some of our recent work resulting in more robust algorithms and a better control of simulation error. Particular emphasis will be given to the important role the density-functional toolkit (DFTK), our in-house Julia-based density functional theory code, has played for facilitating the interactions between mathematicians, computer scientists and application researchers.

Bio:
Michael Herbst obtained a PhD in Theoretical Chemistry from Heidelberg University in 2018, after which he moved on to two postdoctoral research stays in Applied Mathematics. From 2019 till 2021 he worked with Éric Cancès (École des Ponts, Paris, France) and from 2021 till 2023 he stayed in the group of Benjamin Stamm (RWTH Aachen). Since March 2023 he is an assistant professor in the Institute of Mathematics and the Institute of Materials at EPFL. His current research spans broadly in the field of materials simulations concerning numerical error control and uncertainty quantification of DFT models as well as the development of efficient and robust algorithms for high-throughput screening.

More on our website: https://www.epfl.ch/research/domains/...