Torpid Mixing of Markov Chains for the Six-vertex Model on $\mathbb{Z}^2$
Tianyu Liu

TL;DR
This paper proves that certain Markov chain algorithms for the six-vertex model on finite square lattices mix slowly in specific parameter regimes, highlighting limitations in their efficiency for these phases.
Contribution
It is the first to establish torpid mixing of Glauber dynamics and the directed-loop algorithm for the six-vertex model in ferroelectric and anti-ferroelectric phases.
Findings
Markov chains are torpidly mixing in ferroelectric phase
Markov chains are torpidly mixing in anti-ferroelectric phase
Results apply to finite regions of the square lattice
Abstract
In this paper, we study the mixing time of two widely used Markov chain algorithms for the six-vertex model, Glauber dynamics and the directed-loop algorithm, on the square lattice . We prove, for the first time that, on finite regions of the square lattice these Markov chains are torpidly mixing under parameter settings in the ferroelectric phase and the anti-ferroelectric phase.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsMarkov Chains and Monte Carlo Methods · Random Matrices and Applications · Theoretical and Computational Physics
