Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2401.12201 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909079557898240 |
|---|---|
| author | Roberts, Tom Prellberg, Thomas |
| author_facet | Roberts, Tom Prellberg, Thomas |
| contents | Sampling with the Generalised Atmospheric Rosenbluth Method (GARM) is a technique for estimating the distributions of lattice polymer models that has had some success in the study of linear polymers and lattice polygons. In this paper we will explain how and why such sampling appears not to be effective for many models of branched polymers.
Analysing the algorithm on a simple binary tree, we argue that the fundamental issue is an inherent bias towards extreme configurations that is costly to correct with reweighting techniques.
We provide a solution to this by applying uniform sampling methods to the atmospheres that are central to GARM. We caution that the ensuing computational complexity often outweighs the improvements gained. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_12201 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Improving Convergence of Generalised Rosenbluth Sampling for Branched Polymer Models by Uniform Sampling Roberts, Tom Prellberg, Thomas Statistical Mechanics Sampling with the Generalised Atmospheric Rosenbluth Method (GARM) is a technique for estimating the distributions of lattice polymer models that has had some success in the study of linear polymers and lattice polygons. In this paper we will explain how and why such sampling appears not to be effective for many models of branched polymers. Analysing the algorithm on a simple binary tree, we argue that the fundamental issue is an inherent bias towards extreme configurations that is costly to correct with reweighting techniques. We provide a solution to this by applying uniform sampling methods to the atmospheres that are central to GARM. We caution that the ensuing computational complexity often outweighs the improvements gained. |
| title | Improving Convergence of Generalised Rosenbluth Sampling for Branched Polymer Models by Uniform Sampling |
| topic | Statistical Mechanics |
| url | https://arxiv.org/abs/2401.12201 |