Rare event sampling with stochastic growth algorithms

TL;DR

This paper presents advanced stochastic growth algorithms, improving upon the Rosenbluth method, to achieve uniform sampling of complex polymer configurations in statistical mechanics, outperforming previous methods in challenging scenarios.

Contribution

It introduces enhancements to the classic Rosenbluth algorithm, enabling efficient uniform sampling of polymers in complex equilibrium systems where prior algorithms struggled.

Findings

Successfully sampled collapsed polymers near surfaces

Applied to protein folding models with improved results

Demonstrated superiority over existing algorithms in challenging cases

Abstract

We discuss uniform sampling algorithms that are based on stochastic growth methods, using sampling of extreme configurations of polymers in simple lattice models as a motivation. We shall show how a series of clever enhancements to a fifty-odd year old algorithm, the Rosenbluth method, led to a cutting-edge algorithm capable of uniform sampling of equilibrium statistical mechanical systems of polymers in situations where competing algorithms failed to perform well. Examples range from collapsed homo-polymers near sticky surfaces to models of protein folding.