Big Data Meets Quantum Chemistry Approximations: The Δ-Machine Learning Approach
- University of Basel (Switzerland)
- Max-Planck-Institut für Kohlenforschung, Mülheim an der Ruhr (Germany); Friedrich-Alexander University Erlangen-Nuremberg, Bamberg (Germany)
- University of Basel (Switzerland); Argonne National Laboratory (ANL), Argonne, IL (United States). Argonne Leadership Computing Facility (ALCF)
Chemically accurate and comprehensive studies of the virtual space of all possible molecules are severely limited by the computational cost of quantum chemistry. Here we introduce a composite strategy that adds machine learning corrections to computationally inexpensive approximate legacy quantum methods. After training, highly accurate predictions of enthalpies, free energies, entropies, and electron correlation energies are possible, for significantly larger molecular sets than used for training. For thermochemical properties of up to 16k isomers of C7H10O2 we present numerical evidence that chemical accuracy can be reached. We also predict electron correlation energy in post Hartree-Fock methods, at the computational cost of HartreeFock, and we establish a qualitative relationship between molecular entropy and electron correlation. The transferability of our approach is demonstrated, using semiempirical quantum chemistry and machine learning models trained on 1 and 10% of 134k organic molecules, to reproduce enthalpies of all remaining molecules at density functional theory level of accuracy.
- Research Organization:
- Argonne National Laboratory (ANL), Argonne, IL (United States)
- Sponsoring Organization:
- USDOE Office of Science (SC); Swiss National Science Foundation (SNSF)
- Grant/Contract Number:
- AC02-06CH11357; PP00P2_138932
- OSTI ID:
- 1392925
- Journal Information:
- Journal of Chemical Theory and Computation, Vol. 11, Issue 5; ISSN 1549-9618
- Publisher:
- American Chemical SocietyCopyright Statement
- Country of Publication:
- United States
- Language:
- English
Web of Science
Similar Records
Machine Learning of Parameters for Accurate Semiempirical Quantum Chemical Calculations
Deep learning of dynamically responsive chemical Hamiltonians with semiempirical quantum mechanics