Accelerating the Understanding of Rare Events

Extracting mechanistic information is a central challenge in molecular simulations of rare events, from chemical reactions to self-assembly of living matter. The recent increase in machine learning (ML) and artificial intelligence (AI) efforts has completely revolutionized the way that researchers nowadays deal with such rare events. ML methods (neural networks, deep learning, etc.) have been picked up by the rare event community to construct models, optimize force fields, analyze results and even help to accelerate the sampling itself in an adaptive and iterative fashion. Until now the rare-event and the ML communities did not have the chance to come together to exchange recent ideas on these topics. This workshop aims to make an inventory of outstanding problems in the application of statistical mechanics and machine learning (ML) approaches for enhanced rare event sampling and the construction of reliable and meaningful models from atomistic simulation data, in a wide range of fields ranging from physics and chemistry to materials science and molecular biology.