CaltechAUTHORS
  A Caltech Library Service

Multi-task learning for electronic structure to predict and explore molecular potential energy surfaces

Qiao, Zhuoran and Ding, Feizhi and Welborn, Matthew and Bygrave, Peter J. and Smith, Daniel G. A. and Anandkumar, Animashree and Manby, Frederick R. and Miller, Thomas F., III (2020) Multi-task learning for electronic structure to predict and explore molecular potential energy surfaces. . (Unpublished) https://resolver.caltech.edu/CaltechAUTHORS:20201203-151028849

[img] PDF - Accepted Version
See Usage Policy.

3348Kb

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20201203-151028849

Abstract

We refine the OrbNet model to accurately predict energy, forces, and other response properties for molecules using a graph neural-network architecture based on features from low-cost approximated quantum operators in the symmetry-adapted atomic orbital basis. The model is end-to-end differentiable due to the derivation of analytic gradients for all electronic structure terms, and is shown to be transferable across chemical space due to the use of domain-specific features. The learning efficiency is improved by incorporating physically motivated constraints on the electronic structure through multi-task learning. The model outperforms existing methods on energy prediction tasks for the QM9 dataset and for molecular geometry optimizations on conformer datasets, at a computational cost that is thousand-fold or more reduced compared to conventional quantum-chemistry calculations (such as density functional theory) that offer similar accuracy.


Item Type:Report or Paper (Discussion Paper)
Related URLs:
URLURL TypeDescription
http://arxiv.org/abs/2011.02680arXivDiscussion Paper
ORCID:
AuthorORCID
Qiao, Zhuoran0000-0002-5704-7331
Welborn, Matthew0000-0001-8659-6535
Manby, Frederick R.0000-0001-7611-714X
Miller, Thomas F., III0000-0002-1882-5380
Additional Information:Accepted for presentation at the Machine Learning for Molecules workshop at NeurIPS 2020. Z.Q. acknowledges the graduate research funding from Caltech. T.F.M. and A.A. acknowledge partial support from the Caltech DeLogi fund, and A.A. acknowledges support from a Caltech Bren professorship. The authors gratefully acknowledge NVIDIA, including Abe Stern and Tom Gibbs, for helpful discussions regarding GPU implementations of graph neural networks.
Funders:
Funding AgencyGrant Number
Caltech De Logi FundUNSPECIFIED
Bren Professor of Computing and Mathematical SciencesUNSPECIFIED
Record Number:CaltechAUTHORS:20201203-151028849
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20201203-151028849
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:106900
Collection:CaltechAUTHORS
Deposited By: George Porter
Deposited On:05 Dec 2020 01:51
Last Modified:05 Dec 2020 01:51

Repository Staff Only: item control page