Browse publications on Google Scholar (top-right) ↗
Resume (English only)
Academic Achievements
News and views article in Nature Machine Intelligence: Accelerating molecular dynamics by going with the flow
A new collaboration: Learning radical excited states from sparse data
New 3 yr PostDoc position available in MDIG: Research Fellow in Machine Learning for Materials Design
New preprint: Learning disentangled latent representations facilitates discovery and design of functional materials
New preprint: The carbon cost of materials discovery: Can machine learning really accelerate the discovery of new photovoltaics?
2016, Chem: Computational screening of all stoichiometric inorganic materials
2018, Nature: Machine learning for molecular and materials science
2019, npj Computational Materials: Designing interfaces in energy materials applications with first-principles calculations
2022, npj Computational Materials: Distributed representations of atoms and materials for machine learning
2021, The Journal of Chemical Physics: Entropy-based active learning of graph neural network surrogate models for materials properties
2022, Accounts of Materials Research: Interpretable and explainable machine learning for materials science and chemistry
Research Experience
We use a combination of data-driven methods (such as deep learning and Bayesian statistics) and quantum mechanics calculations to design new materials on computers and to help accelerate the experimental characterization of materials. We work with other academics, national facilities, and companies.
Background
The Materials Design and Informatics Group (MDIG) is a research collective, working to accelerate the development of new green energy materials. We are based at UCL, in the Department of Chemistry.