English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Machine learning (ML) models for molecules and materials commonly rely on adecomposition of the global target quantity into local, atom-centeredcontributions. This approach is convenient from a computational perspective,enabling large-scale ML-driven simulations with a linear-scaling cost, and alsoallow for the identification and post-hoc interpretation of contributions fromindividual chemical environments and motifs to complicated macroscopicproperties. However, even though there exist practical justifications for thesedecompositions, only the global quantity is rigorously defined, and thus it isunclear to what extent the atomistic terms predicted by the model can betrusted. Here, we introduce a quantitative metric, which we call the localprediction rigidity (LPR), that allows one to assess how robust the locallydecomposed predictions of ML models are. We investigate the dependence of LPRon the aspects of model training, particularly the composition of trainingdataset, for a range of different problems from simple toy models to realchemical systems. We present strategies to systematically enhance the LPR,which can be used to improve the robustness, interpretability, andtransferability of atomistic ML models.

Robustness of Local Predictions in Atomistic Machine Learning Models