Quantum Energy Regression using Scattering Transforms

TL;DR

This paper introduces a scattering transform-based method for quantum energy regression that achieves state-of-the-art accuracy by leveraging multiscale wavelet transforms with invariant and stable properties.

Contribution

It proposes a novel scattering transform framework for quantum energy prediction, overcoming limitations of previous Coulomb matrix approaches.

Findings

Achieves state-of-the-art accuracy on planar molecules.

Demonstrates robustness and stability of the scattering transform approach.

Outperforms traditional Coulomb matrix methods.

Abstract

We present a novel approach to the regression of quantum mechanical energies based on a scattering transform of an intermediate electron density representation. A scattering transform is a deep convolution network computed with a cascade of multiscale wavelet transforms. It possesses appropriate invariant and stability properties for quantum energy regression. This new framework removes fundamental limitations of Coulomb matrix based energy regressions, and numerical experiments give state-of-the-art accuracy over planar molecules.