DIRECT Sampling for Robust MLPs

Ji’s work on “Robust training of machine learning interatomic potentials with dimensionality reduction and stratified sampling” is now out in npj Computational Materials! Machine learning interatomic potentials (MLIPs) enable accurate simulations of materials at scales beyond that accessible by ab initio methods. In this work, we present DImensionality-Reduced Encoded Clusters with sTratified (DIRECT) sampling as an approach to select a robust training set of structures from a large and complex configuration space. By applying DIRECT sampling on the Materials Project relaxation trajectories dataset with over one million structures and 89 elements, we develop an improved materials 3-body graph network (M3GNet) universal potential that extrapolates more reliably to unseen structures. We further show that molecular dynamics (MD) simulations with the M3GNet universal potential can be used instead of expensive ab initio MD to rapidly create a large configuration space for target systems. We combined this scheme with DIRECT sampling to develop a reliable moment tensor potential for titanium hydrides without the need for iterative augmentation of training structures. Check out this work here. If you want to use DIRECT sampling for your work, please check out our implementation available on our MAML repository on Github.

MRS Fall 2023 Tutorial on ML for SSBs

Ji Qi gave a tutorial talk on “Machine Learning and High-Throughput Discovery and Design of Next Generation Electrode and Superionic Materials and Their Interfaces for SSBs” at the MRS Fall 2023! This tutorial provides an overview of how our group is using ML techniques to gain insights and discovery alkali superionic conductors, as well as the many open-source software packages that we have developed for these purposes. A recording of this talk is available on our group’s YouTube channel (and embedded above).