Professor of Molecular Modelling
Academic Division: Mechanics, Materials and Design
Research group: Applied Mechanics
Telephone: +44 1223 7 66966
My expertise is in atomistic simulation, particularly in multi scale modelling that couples quantum mechanics to larger length scales. I am currently engaged in applying machine learning and other data intensive techniques to materials modelling problems, such as deriving force fields (interatomic potentials) from ab initio electronic structure data. Also interested in statistical problems in molecular dynamics, e.g. in enhanced sampling algorithms that can be used explore the global configuration space of materials and molecules.
I help run an informal meeting called Machine Learning Discussion Group (MLDG) where we discuss the application of machine learning to physics, chemistry and materials science problems. You can subscribe to the MLDG mailing list (with a current Cambridge network ID), and get information about the talks.
Audio and Video
"Machine learning the quantum mechanics of materials and molecules" in July 2020 at the Ellis workshop on "Quantum and physics based machine learning".
An online seminar from June 2020 on machine learned interatomic potentials, quite technical in content, in the ML4Science series, with the now recording on YouTube.
A seminar at IPAM on machine learning and force fields for materials and molecules, part of the "Machine Learning for Physics and the Physics of Learning" programme in the fall of 2019.
A seminar in Edinburgh at ICMS given in the spring of 2019 on machine learning, materials science, and the Gaussian Approximation Potential (GAP) models.
A podcast by Physics World (my bit starts at around 22:25, 12 minutes long) in which the general idea machine learning in the physical sciences is discussed, along with open access publishing (October 2019).