Many Molecular Properties from One Kernel in Chemical Space

TL;DR

This paper presents a universal kernel approach for machine learning models that can predict multiple molecular properties simultaneously, enabling rapid and scalable property prediction across large chemical datasets.

Contribution

The authors introduce property-independent kernels that encode molecular structures, allowing for the instant creation of models for various properties and systematic improvement with more data.

Findings

Models accurately predict multiple molecular properties.

The kernel approach is scalable to large datasets.

Models can be improved by adding more data.

Abstract

We introduce property-independent kernels for machine learning modeling 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. Corresponding molecular reference properties provided, they enable the instantaneous generation of ML models which can systematically be 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 kilo organic molecules of similar size. Resulting models are discussed as well as the kernels' use for generating and using other property models.