Recommended by

Indexed by

Robustness of local predictions in atomistic machine learning models

Sanggyu Chong1*, Federico Grasselli1, Chiheb Ben Mahmoud1, Joe Morrow2, Volker Deringer2, Michele Ceriotti1

1 Laboratory of Computational Science and Modeling (COSMO), IMX, École Polytechnique Fédérale de Lausanne (EPFL), 1015 Lausanne, Switzerland

2 Department of Chemistry, Inorganic Chemistry Laboratory, University of Oxford, Oxford OX1 3QR, United Kingdom

* Corresponding authors emails: sanggyu.chong@epfl.ch
DOI10.24435/materialscloud:re-0d [version v1]

Publication date: Jun 30, 2023

How to cite this record

Sanggyu Chong, Federico Grasselli, Chiheb Ben Mahmoud, Joe Morrow, Volker Deringer, Michele Ceriotti, Robustness of local predictions in atomistic machine learning models, Materials Cloud Archive 2023.102 (2023), doi: 10.24435/materialscloud:re-0d.


Machine learning (ML) models for molecules and materials commonly rely on a decomposition of the global target quantity into local, atom-centered contributions. This approach is convenient from a computational perspective, enabling large-scale ML-driven simulations with a linear-scaling cost, and can also be used to deduce useful structure--property relations as they associate simple atomic motifs with complicated macroscopic properties. However, even though there exist practical justifications for these decompositions, only the global quantity is rigorously defined, and thus it is unclear to what extent the atomistic terms predicted by the model can be trusted. Here, we introduce a quantitative metric, which we call the local prediction rigidity (LPR), to assess how robust the locally decomposed predictions of ML models are. We investigate the dependence of LPR on the details of model training, e.g. composition of the dataset, for several different problems ranging from simple toy models to real chemical systems. We present strategies to systematically enhance the LPR, which can be used to improve the robustness, interpretability, and transferability of the resulting atomistic ML models. This repository contains datasets and Jupyter notebooks employed to corroborate the results of the above study.

Materials Cloud sections using this data

No Explore or Discover sections associated with this archive record.


File name Size Description
12.9 MiB tar file containing datasets and notebooks
1.8 KiB README file containing instructions to run the notebooks correctly


Files and data are licensed under the terms of the following license: Creative Commons Attribution 4.0 International.
Metadata, except for email addresses, are licensed under the Creative Commons Attribution Share-Alike 4.0 International license.


Marie Curie Fellowship ERC EPSRC

Version history:

2023.102 (version v1) [This version] Jun 30, 2023 DOI10.24435/materialscloud:re-0d