Beyond the locality approximation in the standard diffusion Monte Carlo method

TL;DR

This paper introduces a novel approach to incorporate non-local potentials into the standard diffusion Monte Carlo method without relying on the locality approximation, enhancing stability and accuracy.

Contribution

It proposes a stochastic projection method based on a fixed node effective Hamiltonian that maintains variational properties with non-local operators.

Findings

Provides a stable diffusion process for divergent non-local potentials

Achieves an upper bound on the true ground state energy

Simplifies the modification needed for the standard DMC algorithm

Abstract

We present a way to include non local potentials in the standard Diffusion Monte Carlo method without using the locality approximation. We define a stochastic projection based on a fixed node effective Hamiltonian, whose lowest energy is an upper bound of the true ground state energy, even in the presence of non local operators in the Hamiltonian. The variational property of the resulting algorithm provides a stable diffusion process, even in the case of divergent non local potentials, like the hard-core pseudopotentials. It turns out that the modification required to improve the standard Diffusion Monte Carlo algorithm is simple.