A Simple and Scalable Kernel Density Approach for Reliable Uncertainty Quantification in Atomistic Machine Learning

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11310%2F25%3A10505098" target="_blank" >RIV/00216208:11310/25:10505098 - isvavai.cz</a>

Výsledek na webu

<a href="https://verso.is.cuni.cz/pub/verso.fpl?fname=obd_publikace_handle&handle=5CM7Gt3TtR" target="_blank" >https://verso.is.cuni.cz/pub/verso.fpl?fname=obd_publikace_handle&handle=5CM7Gt3TtR</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1021/acs.jpclett.5c02595" target="_blank" >10.1021/acs.jpclett.5c02595</a>

Jazyk výsledku

angličtina

Název v původním jazyce

A Simple and Scalable Kernel Density Approach for Reliable Uncertainty Quantification in Atomistic Machine Learning

Popis výsledku v původním jazyce

Machine learning models are increasingly used to predict material properties and accelerate atomistic simulations, but the reliability of their predictions depends on the representativeness of the training data. We present a scalable, GPU-accelerated uncertainty quantification framework based on k-nearest-neighbor kernel density estimation (KDE) in a PCA-reduced descriptor space. This method efficiently detects sparsely sampled regions in large, high-dimensional data sets and provides a transferable, model-agnostic uncertainty metric without requiring retraining costly model ensembles. The framework is validated across diverse case studies varying in (i) chemistry, (ii) prediction models (including foundational neural network), (iii) descriptors used for KDE estimation, and (iv) properties whose uncertainty is sought. In all cases, the KDE-based score reliably flags extrapolative configurations, correlates well with conventional ensemble-based uncertainties, and highlights regions of reduced prediction trustworthiness. The approach offers a practical route for improving the interpretability, robustness, and deployment readiness of ML models in materials science.

Název v anglickém jazyce

A Simple and Scalable Kernel Density Approach for Reliable Uncertainty Quantification in Atomistic Machine Learning

Popis výsledku anglicky

Machine learning models are increasingly used to predict material properties and accelerate atomistic simulations, but the reliability of their predictions depends on the representativeness of the training data. We present a scalable, GPU-accelerated uncertainty quantification framework based on k-nearest-neighbor kernel density estimation (KDE) in a PCA-reduced descriptor space. This method efficiently detects sparsely sampled regions in large, high-dimensional data sets and provides a transferable, model-agnostic uncertainty metric without requiring retraining costly model ensembles. The framework is validated across diverse case studies varying in (i) chemistry, (ii) prediction models (including foundational neural network), (iii) descriptors used for KDE estimation, and (iv) properties whose uncertainty is sought. In all cases, the KDE-based score reliably flags extrapolative configurations, correlates well with conventional ensemble-based uncertainties, and highlights regions of reduced prediction trustworthiness. The approach offers a practical route for improving the interpretability, robustness, and deployment readiness of ML models in materials science.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

10403 - Physical chemistry

Projekt

Výsledek vznikl pri realizaci vícero projektů. Více informací v záložce Projekty.

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2025

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

Journal of Physical Chemistry Letters

ISSN

1948-7185

e-ISSN

1948-7185

Svazek periodika

16

Číslo periodika v rámci svazku

42

Stát vydavatele periodika

US - Spojené státy americké

Počet stran výsledku

6

Strana od-do

11081-11086

Kód UT WoS článku

001595343700001

EID výsledku v databázi Scopus

2-s2.0-105018905942