[Paper Review] Improved Sweeping Cluster Algorithm for Quantum Dimer Model

Zheng Yan|arXiv (Cornell University)|Nov 16, 2020

Quantum many-body systems1 references4 citations

TL;DR

This paper proposes an improved sweeping cluster algorithm for quantum dimer models (QDMs) that extends sampling across multiple winding (topological) sectors, overcoming a key limitation of prior methods. By enabling efficient cluster updates in all sectors, the method enhances the efficiency and accuracy of world-line quantum Monte Carlo simulations, particularly away from the Rokhsar-Kivelson point where projector QMC fails.

ABSTRACT

Quantum dimer models~(QDMs) featured by strong geometric constraint are effective low energy descriptions of many quantum spin systems. The geometric restriction described by local gauge field, hinders the application of numerical algorithms. Before sweeping cluster method was applied in world-line quantum Monte Carlo (QMC) algorithm, there is only projector QMC which obey the constraints and could be used for calculation on QDMs. However, the projector QMC for QDMs has some drawbacks, e.g., it is not effective when the parameter interval away from Rokhsar-Kivelson (RK) point. That's because the projector method still lacks a cluster update to improve its efficiency. Although sweeping cluster algorithm improves the update for these systems, it also only works in one winding (topological) sector. In this paper, we improve the sweeping cluster method to sample in different winding sectors.

Motivation & Objective

• To address the inefficiency of projector quantum Monte Carlo (QMC) in quantum dimer models (QDMs) away from the Rokhsar-Kivelson (RK) point.
• To overcome the limitation of existing sweeping cluster algorithms, which are restricted to a single winding sector.
• To develop a cluster update method that enables efficient sampling across multiple topological sectors in QDMs.
• To improve the performance and convergence of world-line QMC simulations for strongly constrained quantum systems.

Proposed method

• The method extends the sweeping cluster algorithm to allow updates that cross different winding sectors in the world-line configuration space.
• It introduces a modified cluster update procedure that respects the geometric constraints of the QDM while enabling transitions between sectors.
• The algorithm maintains detailed balance and global balance by carefully constructing cluster moves that preserve the topological winding number distribution.
• The approach integrates with world-line QMC by adapting the cluster growth process to include sector-changing moves.
• The method ensures ergodicity across all winding sectors by allowing clusters to wrap around the system in different topological configurations.
• It uses a modified acceptance criterion to maintain detailed balance during sector transitions, ensuring correct statistical sampling.

Experimental results

Research questions

• RQ1 Can sweeping cluster updates be generalized to sample across multiple winding sectors in quantum dimer models?
• RQ2 How can cluster updates be designed to maintain detailed balance while transitioning between different topological sectors?
• RQ3 What is the impact of multi-sector sampling on the efficiency and convergence of world-line QMC simulations for QDMs?
• RQ4 How does the performance of the improved algorithm compare to projector QMC away from the Rokhsar-Kivelson point?
• RQ5 Can the new method maintain ergodicity and correct statistical sampling in strongly constrained quantum systems?

Key findings

• The improved sweeping cluster algorithm successfully enables sampling across multiple winding sectors in quantum dimer models.
• The method enhances simulation efficiency, particularly in parameter regimes far from the Rokhsar-Kivelson point where projector QMC fails.
• The algorithm maintains detailed balance and global balance during sector transitions, ensuring correct statistical sampling.
• The extension allows for more accurate and reliable simulations of QDMs in diverse physical regimes.
• The approach provides a viable alternative to projector QMC for strongly constrained quantum systems with improved convergence and robustness.