Recorded 25 January 2023. Stefan Chmiela of the Technische Universität Berlin, Machine Learning, presents "Accurate global machine learning force fields for molecules with hundreds of atoms" at IPAM's Learning and Emergence in Molecular Systems Workshop. Abstract: Global machine learning force fields (MLFFs), that have the capacity to capture collective many-atom interactions in molecular systems, currently only scale up to a few dozen atoms due a considerable growth of model complexity with system size. For larger molecules, locality assumptions are typically introduced, with the consequence that non-local interactions are poorly or not at all described, even if those interactions are contained within the reference ab initio data. Some recent studies indicate that this can lead to an inaccurate representations of observables from molecular dynamics simulations due to error accumulation. We approach this challenge and review several research directions towards reconstructing accurate global MLFFs for systems with up to several hundred atoms, without resorting to any localization of atomic interactions or other potentially uncontrolled approximations. Learn more online at: http://www.ipam.ucla.edu/programs/workshops/learning-and-emergence-in-molecular-systems/