TY - GEN AU - Oliver T. Unke AU - Stefan Chmiela AU - Huziel E. Sauceda AU - Michael Gastegger AU - Igor Poltavsky AU - Kristof T. Schütt AU - Alexandre Tkatchenko AU - Klaus-Robert Müller TI - Machine Learning Force Fields PY - 2021/// CY - [Place of publication not identified] PB - American Chemical Society KW - machine learning interatomic potentials KW - potential energy surface KW - chemical bonding KW - interatomic interactions KW - molecular dynamics simulation KW - B. Sc. Artificial Intelligence KW - UMUN open access collection KW - machine learning KW - Review articles N1 - Publication type: Review article; Linked open-access copy: published version; Open access; no subscription required for the linked open-access version N2 - methods and the efficiency of classical FFs. The key idea is to learn the statistical relation between chemical structure and potential energy without relying on a preconceived notion of fixed chemical bonds or knowledge about the relevant interactions. Such universal ML approximations are in principle only limited by the quality and quantity of the reference data used to train them. This review gives an overview of applications of ML-FFs and the chemical insights that can be obtained from them. The core concepts underlying ML-FFs are described in detail, and a step-by-step guide for constructing and testing them from scratch is given. The text concludes with a discussion of the challenges that remain to be overcome by the next generation of ML-FFs UR - https://arxiv.org/pdf/2010.07067 ER -