Solving the Schr\"odinger Equation in the Configuration Space with Generative Machine Learning

TL;DR

This paper introduces a machine learning method that uses a generative model to efficiently select important configurations, significantly accelerating the solution of the Schrödinger equation for molecules and materials.

Contribution

The paper presents a novel iterative generative machine learning approach for configuration selection in electronic structure calculations, improving convergence speed.

Findings

Achieves chemical accuracy faster than random sampling

Reduces computational cost in configuration interaction calculations

Demonstrates effectiveness on molecular applications

Abstract

The configuration interaction approach provides a conceptually simple and powerful approach to solve the Schr\"odinger equation for realistic molecules and materials but is characterized by an unfavourable scaling, which strongly limits its practical applicability. Effectively selecting only the configurations that actually contribute to the wavefunction is a fundamental step towards practical applications. We propose a machine learning approach that iteratively trains a generative model to preferentially generate the important configurations. By considering molecular applications it is shown that convergence to chemical accuracy can be achieved much more rapidly with respect to random sampling or the Monte Carlo configuration interaction method. This work paves the way to a broader use of generative models to solve the electronic structure problem.