Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
August 18, 2025Open Access

Generalized Machine Learning Potential Models for Elemental Nanoclusters

View Full Paper
Ask AI
Bookmark
Share

Authors

SSSubramanian K. R. S. SankaranarayananArgonne National LaboratorySBSuvo BanikArgonne National LaboratoryAAAbhishek AggarwalAmazon (Germany)

Discussion

Loading...

Member takes

Implication

This research develops machine learning potential models to accurately predict the properties of nanoclusters, highlighting their significance in catalysis and energy storage.

Key Points

  • The GAP models achieve strong agreement with density functional theory in energy predictions, showcasing high accuracy.
  • Over 170,000 nanocluster configurations were utilized, demonstrating extensive training for reliable predictions.
  • The modeling approach combines structural weighting with a Bayesian training method, enhancing generalization capabilities.
  • This work offers a promising step towards universal interatomic potentials with ab-initio fidelity across diverse elemental nanoclusters.

Cite This Study

Sankaranarayanan et al. (2025) studied this question.

synapsesocial.com/papers/68af453fad7bf08b1ead2baahttps://doi.org/10.21203/rs.3.rs-7106351/v1
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Development of a Machine Learning Potential to Study Structure and Thermodynamics of Nickel Nanoclusters2024 · 1 citations
  2. 2Fast and accurate machine-learned interatomic potentials for large-scale simulations of Cu, Al and Ni2024 · 2 citations
  3. 3Cluster Models for Next‐Generation, Machine‐Learning‐Based Energy Functions for Molecular Simulations2026
  4. 4Active and Transfer Learning of High-Dimensional Neural Network Potentials for Transition Metals2024 · 8 citations
  5. 5Machine Learning Interatomic Potentials for Energy Materials: Architectures, Training Strategies, and Applications2026 · 12 citations