English

Generalized Random Energy Model at Complex Temperatures

Probability 2014-02-12 v1 Disordered Systems and Neural Networks Mathematical Physics Complex Variables math.MP

Abstract

Motivated by the Lee--Yang approach to phase transitions, we study the partition function of the Generalized Random Energy Model (GREM) at complex inverse temperature β\beta. We compute the limiting log-partition function and describe the fluctuations of the partition function. For the GREM with dd levels, in total, there are 12(d+1)(d+2)\frac 12 (d+1)(d+2) phases, each of which can symbolically be encoded as Gd1Fd2Ed3G^{d_1}F^{d_2}E^{d_3} with d1,d2,d3N0d_1,d_2,d_3\in\mathbb{N}_0 such that d1+d2+d3=dd_1+d_2+d_3=d. In phase Gd1Fd2Ed3G^{d_1}F^{d_2}E^{d_3}, the first d1d_1 levels (counting from the root of the GREM tree) are in the glassy phase (G), the next d2d_2 levels are dominated by fluctuations (F), and the last d3d_3 levels are dominated by the expectation (E). Only the phases of the form Gd1Ed3G^{d_1}E^{d_3} intersect the real β\beta axis. We describe the limiting distribution of the zeros of the partition function in the complex β\beta plane (= Fisher zeros). It turns out that the complex zeros densely touch the positive real axis at dd points at which the GREM is known to undergo phase transitions. Our results confirm rigorously and considerably extend the replica-method predictions from the physics literature.

Keywords

Cite

@article{arxiv.1402.2142,
  title  = {Generalized Random Energy Model at Complex Temperatures},
  author = {Zakhar Kabluchko and Anton Klimovsky},
  journal= {arXiv preprint arXiv:1402.2142},
  year   = {2014}
}

Comments

109 pages, 10 figures

R2 v1 2026-06-22T03:04:47.780Z