Learning molecular energies using localized graph kernels
This work addresses the problem of efficiently modeling molecular energies for computational chemistry, though it appears incremental as it builds on existing graph kernel approaches.
The authors tackled the challenge of incorporating physical symmetries into machine learning models for molecular energy prediction by introducing a graph-based method that naturally encodes translation, rotation, and permutation invariances, achieving chemical-level accuracy on a standard dataset of organic molecules.
Recent machine learning methods make it possible to model potential energy of atomic configurations with chemical-level accuracy (as calculated from ab-initio calculations) and at speeds suitable for molecular dynam- ics simulation. Best performance is achieved when the known physical constraints are encoded in the machine learning models. For example, the atomic energy is invariant under global translations and rotations, it is also invariant to permutations of same-species atoms. Although simple to state, these symmetries are complicated to encode into machine learning algorithms. In this paper, we present a machine learning approach based on graph theory that naturally incorporates translation, rotation, and permutation symmetries. Specifically, we use a random walk graph kernel to measure the similarity of two adjacency matrices, each of which represents a local atomic environment. This Graph Approximated Energy (GRAPE) approach is flexible and admits many possible extensions. We benchmark a simple version of GRAPE by predicting atomization energies on a standard dataset of organic molecules.