Goldstandard solutions to the Schrödinger equation using deep learning: How much physics do we need?
Abstract
Finding accurate solutions to the Schrödinger equation is the key unsolved challenge of computational chemistry. Given its importance for the development of new chemical compounds, decades of research have been dedicated to this problem, but due to the large dimensionality even the best available methods do not yet reach the desired accuracy. Recently the combination of deep learning with Monte Carlo methods has emerged as a promising way to obtain highly accurate energies and moderate scaling of computational cost. In this paper we significantly contribute towards this goal by introducing a novel deeplearning architecture that achieves 4070% lower energy error at 6x lower computational cost compared to previous approaches. Using our method we establish a new benchmark by calculating the most accurate variational ground state energies ever published for a number of different atoms and molecules. We systematically break down and measure our improvements, focusing in particular on the effect of increasing physical prior knowledge. We surprisingly find that increasing the prior knowledge given to the architecture can actually decrease accuracy.
 Publication:

arXiv eprints
 Pub Date:
 May 2022
 DOI:
 10.48550/arXiv.2205.09438
 arXiv:
 arXiv:2205.09438
 Bibcode:
 2022arXiv220509438G
 Keywords:

 Computer Science  Machine Learning;
 Physics  Chemical Physics;
 Physics  Computational Physics
 EPrint:
 10 pages + apppendix, 7 figures