Quantum Energy Regression using Scattering Transforms

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.

9 pages, 2 figures, 1 table. v2: Correction to Section 4.3. v3: Replaced by arXiv:1605.04654

References in corpus (1)

• On representing chemical environments

Cited by in corpus (8)

• Moment Tensor Potentials: a class of systematically improvable interatomic potentials
• Quantum-Chemical Insights from Deep Tensor Neural Networks
• Machine Learning of Accurate Energy-Conserving Molecular Force Fields
• Understanding Deep Convolutional Networks
• Analysis of time-frequency scattering transforms
• Size and Temperature Transferability of Direct and Local Deep Neural Networks for Atomic Forces
• Constant Size Molecular Descriptors For Use With Machine Learning
• Machine Learning, Quantum Mechanics, and Chemical Compound Space