Seminar: Machine learning and molecular dynamics

October 5^th 2019, CH2 room at 11am

Atom based computer simulation is one of the most important tools of contemporary physical chemistry. In spite of its many successes, it suffers from severe limitations. Here we show how machine-learning techniques can help in solving at least two different problems. The first one is the accuracy of current interatomic potential models; the second is the limited time scale that simulations can explore. In order to solve the first problem we train a neural network on a set of accurate but expensive quantum chemical calculations. In this way, it is possible to obtain an accurate description of the system at a relatively low computational cost. Crucial for the success of this program has been the design of the neural work and the selection of the training set. We apply this approach to study a metal non-metal transition and to chemical reactions in condensed phases. These applications would not have been possible without the use of efficient sampling methods capable of lifting the time scale barrier. To this effect, we have developed two very efficient sampling methods, metadynamics and variationally enhanced sampling. Both methods are based on the identification of appropriate collective variables, or slow modes, whose sampling needs to be accelerated. Machine learning can be used also for the construction of efficient collective variables based on a modification of the well-known linear discriminant analysis classification method. Finally, we use the variational enhanced sampling approach and a deep neural network to further increase our sampling ability.

Bio: Professor Parrinello is known for his many technical innovations in the field ofatomistic simulations and for a wealth of interdisciplinary applications rangingfrom materials science to chemistry and biology. Together with Roberto Car heintroduced ab-initio molecular dynamics, also known as the Car-Parrinello method, marking the beginning of a new era both in the area of electronic structure calculations and in molecular dynamics simulations. He is also known for the Parrinello-Rahman method, which allows crystalline phase transitionsto be studied by molecular dynamics. More recently he has introduced metadynamics for the study of rare events and the calculation of free energies. Forhis work he has been awarded many prizes and honorary degrees. He is amember of numerous academies and learned societies, including the German Berlin-Brandenburgische Akademie der Wissenschaften, the British Royal Societyand the Italian Accademia Nazionale dei Lincei, which is the major academyin his home country Italy. Born in Messina in 1945, he got his degree from the University of Bologna and is currently professor of Computational Sciencesat Università della Svizzera italiana (Faculty of Informatics) and ETHZ in Switzerland.