Three members of the Maurer Group, Wojciech Stark, Joe Gilkes, and Shay Chaudhuri, will be presenting "An Introduction to Machine Learning for Computational Chemistry".

Abstract:

First-principles methods such as density functional theory (DFT) can provide an accurate and reliable description of chemical systems. However, with the increasing complexity and time scales of studied systems, calculating properties and simulating dynamics using these methods quickly becomes computationally infeasible. Machine learning (ML) methods are therefore becoming the state-of-the-art, providing accuracy comparable to the first-principles methods with exceptional computational efficiency. In this talk, we will provide an introduction into ML concepts and introduce some of the ML methods used for chemical systems, with a focus on exploring the ways that both supervised and unsupervised learning methods can be powerful tools for computational chemistry. We will showcase three case studies from the Maurer Group: ML potentials for structure searches and optimisations of hybrid organic-inorganic interfaces, dynamics on surfaces to high throughput targeted molecular design.