Active Learning of Uniformly Accurate Deep Potential Models for Multicomponent Systems

ORAL

Abstract

We propose an active learning procedure called Deep Potential Generator (DP-GEN) for the construction of accurate and transferable potential energy surface (PES) models. This procedure has three major components: exploration, labeling, and training. As an important application, we use DP-GEN to generate an ab-initio trained reactive force field for water that describes both the molecular and the dissociated water phases.

*The work of L.Z. and W.E is supported in part by Major Program of NNSFC under grant 91130005, ONR grant N00014-13-1-0338 and NSFC grant U1430237.
The work of L.Z. and R.C. is supported in part by the DOE with Award Number DE-SC0019394.
The work of H.W. is supported by the National Science Foundation of China under Grants 11501039, 11871110 and 91530322, and the National Key Research and Development Program of China under Grants 2016YFB0201200 and 2016YFB0201203.
The work of D.Y.L. and H.W. is supported by the Science Challenge Project No.~JCKY2016212A502.

–

Presenters

  • Linfeng Zhang

    • Princeton University

Authors

  • Linfeng Zhang

    • Princeton University

  • De-Ye Lin

    • Institute of Applied Physics and Computational Mathematics

  • Han Wang

    • Laboratory of Computational Physics
    • Institute of Applied Physics and Computational Mathematics

  • Roberto Car

    • Princeton University
    • Chemistry, Princeton University

  • Weinan E

    • Princeton University