Recent years have seen a continuous growth of advanced hardware and software performance such that, today, high-throughput studies enable the rational design of novel molecules and materials. However, even with state-of-the-art techniques, the spatio-temporal scales giving rise to material properties necessitate the use of multi-scale simulation techniques, from ab initio quantum mechanics via all-atom and coarse-grain molecular dynamics up to the continuum level. The vast space of model parameters resulting thereof, as well as the large amount of chemical building blocks call for the use of smart methods to deal with the big amount of data that either goes into such models or that results from the predictions. Here, the use of machine learning (ML) techniques (including kernel and deep neural networks) and methods borrowed from the realm of artificial intelligence (AI) (neuromorphic computing and robotic technologies) are needed to advance the field. The goal of this focus session is to connect several disciplines and to bring together top researchers in the fields of molecular simulation, ML, and AI to develop methodology covering the whole simulation pipeline in the aim of advancing the development of novel molecules for efficient neuromorphic computers, improving photovoltaics, enhancing catalyst design and enabling biomedical applications.
Check the website for the exact date and time.
