NewtonNet: a Newtonian message passing network for deep learning of interatomic potentials and forces. (10th May 2022)

• Record Type: Journal Article
• Authors: Haghighatlari, Mojtaba; Li, Jie; Guan, Xingyi; Zhang, Oufan; Das, Akshaya; Stein, Christopher J.; Heidar-Zadeh, Farnaz; Liu, Meili; Head-Gordon, Martin; Bertels, Luke; Hao, Hongxia; Leven, Itai; Head-Gordon, Teresa
• Is Part Of: Digital discovery. Volume 1:Number 3(2022)
• Page Start: 333
• Page End: 343
• Publication Date: 2022-05-10
• Subjects: Chemistry -- Data processing -- Periodicals; Medical sciences -- Data processing -- Periodicals; Machine learning -- Periodicals; 542.85
• DOI: 10.1039/d2dd00008c
• Languages: English
• ISSNs: 2635-098X

Abstract

Abstract : We report a new deep learning message passing network that takes inspiration from Newton's equations of motion to learn interatomic potentials and forces. Abstract : We report a new deep learning message passing network that takes inspiration from Newton's equations of motion to learn interatomic potentials and forces. With the advantage of directional information from trainable force vectors, and physics-infused operators that are inspired by Newtonian physics, the entire model remains rotationally equivariant, and many-body interactions are inferred by more interpretable physical features. We test NewtonNet on the prediction of several reactive and non-reactive high quality ab initio data sets including single small molecules, a large set of chemically diverse molecules, and methane and hydrogen combustion reactions, achieving state-of-the-art test performance on energies and forces with far greater data and computational efficiency than other deep learning models.