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 Plus annual

Presents the access of premium content as premium feature

Premium Content

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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

With massive amounts of atomic simulation data available, there is a hugeopportunity to develop fast and accurate machine learning models to approximateexpensive physics-based calculations. The key quantity to estimate is atomicforces, where the state-of-the-art Graph Neural Networks (GNNs) explicitlyenforce basic physical constraints such as rotation-covariance. However, tostrictly satisfy the physical constraints, existing models have to maketradeoffs between computational efficiency and model expressiveness. Here weexplore an alternative approach. By not imposing explicit physical constraints,we can flexibly design expressive models while maintaining their computationalefficiency. Physical constraints are implicitly imposed by training the modelsusing physics-based data augmentation. To evaluate the approach, we carefullydesign a scalable and expressive GNN model, ForceNet, and apply it to OC20(Chanussot et al., 2020), an unprecedentedly-large dataset of quantum physicscalculations. Our proposed ForceNet is able to predict atomic forces moreaccurately than state-of-the-art physics-based GNNs while being faster both intraining and inference. Overall, our promising and counter-intuitive resultsopen up an exciting avenue for future research.

ForceNet: A Graph Neural Network for Large-Scale Quantum Calculations