Quantum Energy Regression using Scattering Transforms

By Matthew J. Hirn, Nicolas Poilvert, Stéphane Mallat. 2015

31 citations recorded by the source corpus.

We present a novel approach to the regression of quantum mechanical energies based on a scatter- ing transform of an intermediate electron den- sity representation. A scattering transform is a deep convolution network computed with a cas- cade of multiscale wavelet transforms. It pos- sesses appropriate invariant and stability prop- erties for quantum energy regression. This new framework removes fundamental limitations of Coulomb matrix based energy regressions, and numerical experiments give state-of-the-art accu- racy over planar molecules.

Subfields: Atomic and Molecular Physics, and Optics, Materials Chemistry

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Atomic and Molecular Physics, and Optics

In Materials Chemistry

Citations in the corpus

Automatic Chemical Design Using a Data-Driven Continuous Representation of Molecules (2016, background): “Coulomb matrices, [21] scattering transforms [22], atomic distances [23], etc. are based on 3D geometries, usually obtained at an aﬀordable level of theory.”
Understanding deep convolutional networks (2016, background): “For example, quantum molecular energies f (x) are well estimated by sparse regressions over such scattering representations [30].”
Moment Tensor Potentials: A Class of Systematically Improvable Interatomic Potentials (2015, background): “Additionally, in a recent work [19] the authors put forward an approach of regressing the interatomic interaction energy together with the electron density.”