Boosting Quantum Machine Learning Models with a Multilevel Combination Technique: Pople Diagrams Revisited
- 5 December 2018
- journal article
- research article
- Published by American Chemical Society (ACS) in Journal of Chemical Theory and Computation
- Vol. 15 (3) , 1546-1559
Abstract
Inspired by Pople diagrams popular in quantum chemistry, we introduce a hierarchical scheme, based on the multi-level combination (C) technique, to combine various levels of approximations made when calculating molecular energies within quantum chemistry. When combined with quantum machine learning (QML) models, the resulting CQML model is a generalized unified recursive kernel ridge regression which exploits correlations implicitly encoded in training data comprised of multiple levels in multiple dimensions. Here, we have investigated up to three dimensions: Chemical space, basis set, and electron correlation treatment. Numerical results have been obtained for atomization energies of a set of $\sim$7'000 organic molecules with up to 7 atoms (not counting hydrogens) containing CHONFClS, as well as for $\sim$6'000 constitutional isomers of C$_7$H$_{10}$O$_2$. CQML learning curves for atomization energies suggest a dramatic reduction in necessary training samples calculated with the most accurate and costly method. In order to generate milli-second estimates of CCSD(T)/cc-pvdz atomization energies with prediction errors reaching chemical accuracy ($\sim$1 kcal/mol), the CQML model requires only $\sim$100 training instances at CCSD(T)/cc-pvdz level, rather than thousands within conventional QML, while more training molecules are required at lower levels. Our results suggest a possibly favourable trade-off between various hierarchical approximations whose computational cost scales differently with electron number.
Keywords
All Related Versions
Funding Information
- Schweizerischer Nationalfonds zur F?rderung der Wissenschaftlichen Forschung (200021 175747, 407540_167186, PP00P2 138932)
This publication has 76 references indexed in Scilit:
- Combination technique based k-th moment analysis of elliptic problems with random diffusionJournal of Computational Physics, 2013
- Accelerating materials property predictions using machine learningScientific Reports, 2013
- Machine learning of molecular electronic properties in chemical compound spaceNew Journal of Physics, 2013
- A Note on the Construction of L-Fold Sparse Tensor Product SpacesConstructive Approximation, 2013
- Enumeration of 166 Billion Organic Small Molecules in the Chemical Universe Database GDB-17Journal of Chemical Information and Modeling, 2012
- Optimized general sparse grid approximation spaces for operator equationsMathematics of Computation, 2009
- Toward Computational Materials Design: The Impact of Density Functional Theory on Materials ResearchMRS Bulletin, 2006
- Chemical spaceNature, 2004
- Sparse gridsActa Numerica, 2004
- Three-dimensional "Pople diagram"The Journal of Physical Chemistry, 1990