Reweighting in nonequilibrium simulations

TL;DR

This paper introduces a novel reweighting method for nonequilibrium Monte Carlo simulations that enhances efficiency and accuracy without needing stationary distributions, demonstrated on the contact process.

Contribution

A new reweighting scheme applicable out of equilibrium that improves efficiency and precision in studying interacting particle systems.

Findings

Achieved unprecedented precision in critical parameters

Demonstrated efficiency improvement by an order of magnitude

Validated method on contact process in multiple dimensions

Abstract

A simple reweighting scheme is proposed for Monte Carlo simulations of interacting particle systems, permitting one to study various parameter values in a single study, and improving efficiency by an order of magnitude. Unlike earlier reweighting schemes, the present approach does not require knowledge of the stationary probability distribution, and so is applicable out of equilibrium. The method is applied to the contact process in two and three dimensions, yielding the critical parameter and spreading exponents to unprecedented precision.