A hybrid approach to excited-state-specific variational Monte Carlo and doubly excited states

TL;DR

This paper introduces a hybrid variational Monte Carlo method tailored for excited states, demonstrating improved efficiency and accuracy in calculating double excitations in large molecular systems.

Contribution

The work extends a hybrid optimization approach to excited states, showing its effectiveness for double excitations and larger systems beyond traditional methods.

Findings

Superior statistical efficiency over linear method

Good energetic agreement with benchmarks and experiments

Effective treatment of large systems with double excitations

Abstract

We extend our hybrid linear-method/accelerated-descent variational Monte Carlo optimization approach to excited states and investigate its efficacy in double excitations. In addition to showing a superior statistical efficiency when compared to the linear method, our tests on the carbon dimer and cyclopentadiene show good energetic agreement with benchmark methods and experiment, respectively. We also demonstrate the ability to treat double excitations in systems that are too large for a full treatment by selective configuration interaction methods via an application to 4-aminobenzonitrile. Finally, we investigate the stability of state-specific variance optimization against collapse to other states' variance minima and find that symmetry, ansatz quality, and sample size all have roles to play in achieving stability.