Overview articles for ML Interatomic Potentials
ML-based interatomic potentials come in different architectures, such as neural networks and Gaussian processes, each offering unique advantages for different materials and applications. To find out more about the techniques used in this exciting field, as well as the directions in which it is headed, we have prepared a selection of review articles and opinion pieces written by members of our group.
- A review of the future directions for the modelling of disordered materials and the possibility of creating amorphous materials 'by design', powered by ML computational chemistry (Nature Reviews Materials, 2024)
- An opinion piece focusing on the need to improve the nature, quality, and accessibility of structural data as the next challenge for atomistic ML (Nature Computational Science, 2024)
- A tutorial paper outlining different methods of validating the ML potentials both numerically and physically (The Journal of Chemical Physics, 2023)
- A viewpoint discussing the possible directions of the field as computational resources become increasingly powerful (Nature Reviews Materials, 2023)
- A general overview of the possible applications of interatomic potentials in materials science for modelling functional materials (Advanced Materials, 2019)
