Many Molecular Properties from One Kernel in Chemical Space

Abstract

We introduce property-independent kernels for machine learning models of arbitrarily many molecular properties. The kernels encode molecular structures for training sets of varying size, as well as similarity measures sufficiently diffuse in chemical space to sample over all training molecules. When provided with the corresponding molecular reference properties, they enable the instantaneous generation of machine learning models which can be systematically improved through the addition of more data. This idea is exemplified for single kernel based modeling of internal energy, enthalpy, free energy, heat capacity, polarizability, electronic spread, zero-point vibrational energy, energies of frontier orbitals, HOMO-LUMO gap, and the highest fundamental vibrational wavenumber. Models of these properties are trained and tested using 112,000 organic molecules of similar size. The resulting models are discussed as well as the kernels' use for generating and using other property models.