PAWL-Forced Simulated Tempering

TL;DR

This paper demonstrates how the parallel adaptive Wang-Landau algorithm enhances simulated tempering by automating processes and improving efficiency through parallelization, adaptive proposals, and automated bin splitting.

Contribution

It introduces a novel application of the PAWL algorithm to improve and automate simulated tempering methods.

Findings

Parallelization improves efficiency

Adaptive proposals enhance sampling quality

Automated bin splitting simplifies parameter tuning

Abstract

In this short note, we show how the parallel adaptive Wang-Landau (PAWL) algorithm of Bornn et al. (2013) can be used to automate and improve simulated tempering algorithms. While Wang-Landau and other stochastic approximation methods have frequently been applied within the simulated tempering framework, this note demonstrates through a simple example the additional improvements brought about by parallelization, adaptive proposals and automated bin splitting.