arXiv - CS - Computational Complexity Pub Date : 2020-06-26 , DOI: arxiv-2006.14828
Pjotr Buy; Andreas Galanis; Viresh Patel; Guus Regts

We study the computational complexity of approximating the partition function of the ferromagnetic Ising model in the Lee-Yang circle of zeros given by $|\lambda|=1$, where $\lambda$ is the external field of the model. Complex-valued parameters for the Ising model are relevant for quantum circuit computations and phase transitions in statistical physics, but have also been key in the recent deterministic approximation scheme for all $|\lambda|\neq 1$ by Liu, Sinclair, and Srivastava. Here, we focus on the unresolved complexity picture on the unit circle, and on the tantalising question of what happens in the circular arc around $\lambda=1$, where on one hand the classical algorithm of Jerrum and Sinclair gives a randomised approximation scheme on the real axis suggesting tractability, and on the other hand the presence of Lee-Yang zeros alludes to computational hardness. Our main result establishes a sharp computational transition at the point $\lambda=1$; in fact, our techniques apply more generally to the whole unit circle $|\lambda|=1$. We show #P-hardness for approximating the partition function on graphs of maximum degree $\Delta$ when $b$, the edge-interaction parameter, is in the interval $(0,\frac{\Delta-2}{\Delta}]$ and $\lambda$ is a non-real on the unit circle. This result contrasts with known approximation algorithms when $|\lambda|\neq 1$ or $b\in (\frac{\Delta-2}{\Delta},1)$, and shows that the Lee-Yang circle of zeros is computationally intractable, even on bounded-degree graphs.

down
wechat
bug