Molecular dynamics simulations demand an unprecedented combination of accuracy and scalability to tackle grand challenges in catalysis and materials design.To bridge this gap,we present AlphaNet,a local-frame-based eq...Molecular dynamics simulations demand an unprecedented combination of accuracy and scalability to tackle grand challenges in catalysis and materials design.To bridge this gap,we present AlphaNet,a local-frame-based equivariant model that simultaneously improves computational efficiency and predictive precision for interatomic interactions.By constructing equivariant local frames with learnable geometric transitions and enabling contractions through spatial domain and temporal domain,AlphaNet enhances the representational capacity of atomic environments,achieving stateof-the-art accuracy in energy and force predictions.Extensive benchmarks on large-scale datasets spanningmolecular reactions,crystal stability,and surface catalysis(Matbench Discovery and OC2M)demonstrate its superior performance over existing neural network interatomic potentials while ensuring scalability across diverse system sizes with varying types of interatomic interactions.The synergy of accuracy,efficiency,and transferability positions AlphaNet as a transformative tool for modeling multiscale phenomena,decoding dynamics in catalysis and functional interfaces,with direct implications for accelerating the discovery of complex molecular systems and functional materials.Our code and data are available at https://github.com/zmyybc/AlphaNet.展开更多
基金supported by National Key Research and Development Project(2022YFA1503000)National Natural Science Foundation of China(No.92261111)+1 种基金the NSFC Center for Single-Atom Catalysis(No.22388102)We are also grateful to the Center of High-Performance Computing at Tsinghua University for providing computational resources.We also acknowledge the Welch Foundation(F-1841)for support.We thank Prof.Aditi Krishnapriyan for her insightful suggestions and valuable discussions.
文摘Molecular dynamics simulations demand an unprecedented combination of accuracy and scalability to tackle grand challenges in catalysis and materials design.To bridge this gap,we present AlphaNet,a local-frame-based equivariant model that simultaneously improves computational efficiency and predictive precision for interatomic interactions.By constructing equivariant local frames with learnable geometric transitions and enabling contractions through spatial domain and temporal domain,AlphaNet enhances the representational capacity of atomic environments,achieving stateof-the-art accuracy in energy and force predictions.Extensive benchmarks on large-scale datasets spanningmolecular reactions,crystal stability,and surface catalysis(Matbench Discovery and OC2M)demonstrate its superior performance over existing neural network interatomic potentials while ensuring scalability across diverse system sizes with varying types of interatomic interactions.The synergy of accuracy,efficiency,and transferability positions AlphaNet as a transformative tool for modeling multiscale phenomena,decoding dynamics in catalysis and functional interfaces,with direct implications for accelerating the discovery of complex molecular systems and functional materials.Our code and data are available at https://github.com/zmyybc/AlphaNet.
