Wavelet Scattering Regression of Quantum Chemical Energies

TL;DR

This paper presents a wavelet scattering method to estimate quantum chemical energies of molecules, achieving accuracy comparable to DFT with significantly reduced computational cost.

Contribution

It introduces multiscale invariant dictionaries using wavelet scattering for efficient and accurate energy prediction of organic molecules.

Findings

Achieves regression errors similar to DFT.

Reduces computational cost significantly.

Provides state-of-the-art results on molecular databases.

Abstract

We introduce multiscale invariant dictionaries to estimate quantum chemical energies of organic molecules, from training databases. Molecular energies are invariant to isometric atomic displacements, and are Lipschitz continuous to molecular deformations. Similarly to density functional theory (DFT), the molecule is represented by an electronic density function. A multiscale invariant dictionary is calculated with wavelet scattering invariants. It cascades a first wavelet transform which separates scales, with a second wavelet transform which computes interactions across scales. Sparse scattering regressions give state of the art results over two databases of organic planar molecules. On these databases, the regression error is of the order of the error produced by DFT codes, but at a fraction of the computational cost.