ChemRxiv
These are preliminary reports that have not been peer-reviewed. They should not be regarded as conclusive, guide clinical practice/health-related behavior, or be reported in news media as established information. For more information, please see our FAQs.
ccx_manuscript_joint.pdf (5.76 MB)
0/0

Approaching coupled cluster accuracy with a general-purpose neural network potential through transfer learning

preprint
revised on 13.06.2019 and posted on 14.06.2019 by Justin S Smith, Benjamin T. Nebgen, Roman Zubatyuk, Nicholas Lubbers, Christian Devereux, Kipton Barros, Sergei Tretiak, Olexandr Isayev, Adrian Roitberg

Computational modeling of chemical and biological systems at atomic resolution is a crucial tool in the chemist's toolset. The use of computer simulations requires a balance between cost and accuracy: quantum-mechanical methods provide high accuracy but are computationally expensive and scale poorly to large systems, while classical force fields are cheap and scalable, but lack transferability to new systems. Machine learning can be used to achieve the best of both approaches. Here we train a general-purpose neural network potential (ANI-1ccx) that approaches CCSD(T)/CBS accuracy on benchmarks for reaction thermochemistry, isomerization, and drug-like molecular torsions. This is achieved by training a network to DFT data then using transfer learning techniques to retrain on a dataset of gold standard QM calculations (CCSD(T)/CBS) that optimally spans chemical space. The resulting potential is broadly applicable to materials science, biology and chemistry, and billions of times fasterthan CCSD(T)/CBS calculations.

History

Email Address of Submitting Author

roitberg@ufl.edu

Institution

University of Florida

Country

USA

ORCID For Submitting Author

0000-0003-3963-8784

Declaration of Conflict of Interest

None

Version Notes

Final version

Exports