EUVIMEDEUVIMED entdecken
Crossref

Learning Inducing Points and Uncertainty on Molecular Data by Scalable Variational Gaussian Processes

Mikhail Tsitsvero, Mingoo Jin, Andrey Lyalin

SIAM/ASA Journal on Uncertainty Quantification · 30.06.2025 · 13 (2) · 543-562

DOI: 10.1137/23m1549584 ↗

Kurz gefasst

Fachartikel in der Zeitschrift SIAM/ASA Journal on Uncertainty Quantification, online erschienen am 23.04.2025. Verfasst von Mikhail Tsitsvero und 2 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 53 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Uncertainty control and scalability to large datasets are two main issues in the deployment of Gaussian process (GP) models within the autonomous machine learning–based prediction pipelines in material science and chemistry. One way to address both of these issues is by introducing the latent inducing point variables and choosing the right approximation for the marginal log-likelihood objective. Here, we empirically show that variational learning of the inducing points in a molecular descriptor space improves the prediction of energies and atomic forces on two molecular dynamics datasets. First, we show that variational GPs can learn to represent the configurations of the molecules of different types that were not present within the initialization set of configurations. We provide a comparison of alternative log-likelihood training objectives and variational distributions. Among several evaluated approximate marginal log-likelihood objectives, we show that predictive log-likelihood provides excellent uncertainty estimates and good predictive quality. Furthermore, we extend our study to fit the atomic forces for a large molecular crystal system, showing that variational GP models perform well for predicting atomic forces by efficiently learning a sparse representation of the dataset.

Zitieren

APA

Tsitsvero, M., Jin, M., & Lyalin, A. (2025). Learning Inducing Points and Uncertainty on Molecular Data by Scalable Variational Gaussian Processes. SIAM/ASA Journal on Uncertainty Quantification, 13(2), 543-562. https://doi.org/10.1137/23m1549584

Vancouver

Tsitsvero M, Jin M, Lyalin A. Learning Inducing Points and Uncertainty on Molecular Data by Scalable Variational Gaussian Processes. SIAM/ASA Journal on Uncertainty Quantification. 2025;13(2):543-562. doi:10.1137/23m1549584

BibTeX
@article{Tsitsvero2025,
  author = {Tsitsvero, Mikhail and Jin, Mingoo and Lyalin, Andrey},
  title = {Learning Inducing Points and Uncertainty on Molecular Data by Scalable Variational Gaussian Processes},
  journal = {SIAM/ASA Journal on Uncertainty Quantification},
  year = {2025},
  volume = {13},
  number = {2},
  pages = {543--562},
  doi = {10.1137/23m1549584}
}
Kostenlos starten →