Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

AIP conference proceedings

Random field Ising chain and neutral networks with synchronous dynamics

N S Skantzos

Published Jan 1, 2001• 1 citation
Publisher

Abstract

We first present an exact solution of the one-dimensional random-field Ising model in which spin-updates are made fully synchronously, i.e. in parallel (in contrast to the more conventional Glauber-type sequential rules). We find transitions where the support of local observables turns from a continuous interval into a Cantor set and we show that synchronous and sequential random-field models lead asymptotically to the same physical states. We then proceed to an application of these techniques to recurrent neural networks where 1D short-range interactions are combined with infinite-range ones. Due to the competing interactions these models exhibit phase diagrams with first-order transitions and regions with multiple locally stable solutions for the macroscopic order parameters.

DOI
10.1063/1.1358170

Related papers

Glauber Dynamics for One-dimensional Spin Models with Random Fields

Gabor Forgács, David Mukamel, Robert A. Pelcovits

We present an exact solution of the long-time relaxational behavior of the magnetization in the Ising and $\mathrm{XY}$ chains in a quenched random field. The random field is assumed to be of infinite strength but present at only a fraction of the spin sites. We use Glauber dynamics for the Ising case, and a suitably generalized master equation for the $\mathrm{XY}$ case. In both models we find that for $T=0$, as time $t\ensuremath{\rightarrow}\ensuremath{\infty}$ the magnetization decays as exp (${\ensuremath{-}Ct}^{\frac{1}{3}}$), where $C$ is a constant. At finite temperatures the ultimate asymptotic behavior is purely exponential.

Physical review. B, Condensed matter•Jul 1, 1984•23 citations

One-dimensional kinetic Ising model with competing spin-flip and spin-exchange dynamics: Ordering in the case of long-range exchanges

Michel Droz, Zoltàn Ràcz, P. Tartaglia

Ordering of a one-dimensional stochastic Ising model evolving by a combination of spin flips and spin exchanges is investigated. The spin-flip rates satisfy detailed balance for the equilibrium state of the Ising model at temperature T, while the spin exchanges are random and of arbitrary range. Analytical methods and Monte Carlo simulations are used to show that, depending on the details of the spin-flip rate, finite-temperature phase transition may or may not occur in the system. When ordering occurs, it is of mean-field type and the scaling function describing the finite-size scaling of the magnetization fluctuations is found to be indistinguishable from that of an equilibrium Ising model with infinite-range interaction.

Physical Review A•Jun 1, 1990•35 citations

First-order phase transition in a 2D random-field Ising model with conflicting dynamics

Nuno Crokidakis

The effects of locally random magnetic fields are considered in a nonequilibrium Ising model defined on a square lattice with nearest-neighbor interactions. In order to generate the random magnetic fields, we have considered random variables { h } that change randomly with time according to a double-Gaussian probability distribution, which consists of two single-Gaussian distributions, centered at + h o and − h o , with the same width σ. This distribution is very general and can recover in appropriate limits the bimodal distribution ( ) and the single-Gaussian one ( h o = 0). We performed Monte Carlo simulations in lattices with linear sizes in the range L = 32–512. The system exhibits ferromagnetic and paramagnetic steady states. Our results suggest the occurrence of first-order phase transitions between the above-mentioned phases at low temperatures and large random-field intensities h o for some small values of the width σ. By means of finite size scaling, we estimate the critical exponents in the low-field region, where we have continuous phase transitions. In addition, we show a sketch of the phase diagram of the model for some values of σ.

arXiv:0901.2659•Journal of Statistical Mechanics Theory and Experiment•Feb 24, 2009•7 citations

Exact solution of Ising model on a small-world network

J. Viana Lopes, Yu. G. Pogorelov, J. M. B. Lopes dos Santos +1 more

We present an exact solution of a one-dimensional Ising chain with both nearest-neighbor and random long-range interactions. Not surprisingly, the solution confirms the mean-field character of the transition. This solution also predicts the finite-size scaling that we observe in numerical simulations.

arXiv:cond-mat/0402138•Physical Review E•Aug 25, 2004•51 citations

Exact mean-field solution of a spin chain with short-range and long-range interactions

Etienne Granet

We consider the transverse field Ising model with additional all-to-all interactions between the spins. We show that a mean-field treatment of this model becomes exact in the thermodynamic limit, despite the presence of 1D short-range interactions. Namely, we show that the eigenstates of the model are coherent states with an amplitude that varies through the Hilbert space, within which expectation values of local observables can be computed with mean-field theory. We study then the thermodynamics of the model and identify the different phases. Among its peculiar features, this 1D model possesses a second-order phase transition at finite temperature and exhibits inverse melting.

arXiv:2209.08756•SciPost Physics•May 26, 2023•7 citations

Model of a quasi-one-dimensional spin glass

Matej Orešič, R. Pirc

We consider a system of linear chains of Ising spins with short-range nearest-neighbor ferromagnetic interactions of strength K along the chains, and with infinite-range random interchain interactions of a Sherrington-Kirkpatrick type. Also included in the model is a local Gaussian random field with variance \ensuremath{\Delta}. In the replica symmetric case, the system is mapped exactly on the one-dimensional Ising model in a random field and a recursion method for calculating the spin-correlation functions is developed. For \ensuremath{\Delta}=0, an analytic relation for freezing temperature ${\mathit{T}}_{\mathit{f}}$(K) is obtained. The replica-symmetric spin-glass phase is shown to be stable above the freezing temperature ${\mathit{T}}_{\mathit{f}}$(K,\ensuremath{\Delta}), which is determined numerically.

Physical review. B, Condensed matter•Feb 1, 1993•16 citations
We first present an exact solution of the one-dimensional random-field Ising model in which spin-updates are made fully synchronously, i.e. in parallel (in contrast to the more conventional Glauber-type sequential rules). We find transitions where the support of local observables turns from a continuous interval into a Cantor set and we show that synchronous and sequential random-field models lead asymptotically to the same physical states. We then proceed to an application of these techniques to recurrent neural networks where 1D short-range interactions are combined with infinite-range ones. Due to the competing interactions these models exhibit phase diagrams with first-order transitions and regions with multiple locally stable solutions for the macroscopic order parameters.
We present an exact solution of the long-time relaxational behavior of the magnetization in the Ising and XY\mathrm{XY}XY chains in a quenched random field. The random field is assumed to be of infinite strength but present at only a fraction of the spin sites. We use Glauber dynamics for the Ising case, and a suitably generalized master equation for the XY\mathrm{XY}XY case. In both models we find that for T=0T=0T=0, as time $t\ensuremath{\rightarrow}\ensuremath{\infty}$ the magnetization decays as exp (${\ensuremath{-}Ct}^{\frac{1}{3}}$), where CCC is a constant. At finite temperatures the ultimate asymptotic behavior is purely exponential.
Ordering of a one-dimensional stochastic Ising model evolving by a combination of spin flips and spin exchanges is investigated. The spin-flip rates satisfy detailed balance for the equilibrium state of the Ising model at temperature T, while the spin exchanges are random and of arbitrary range. Analytical methods and Monte Carlo simulations are used to show that, depending on the details of the spin-flip rate, finite-temperature phase transition may or may not occur in the system. When ordering occurs, it is of mean-field type and the scaling function describing the finite-size scaling of the magnetization fluctuations is found to be indistinguishable from that of an equilibrium Ising model with infinite-range interaction.
The effects of locally random magnetic fields are considered in a nonequilibrium Ising model defined on a square lattice with nearest-neighbor interactions. In order to generate the random magnetic fields, we have considered random variables { h } that change randomly with time according to a double-Gaussian probability distribution, which consists of two single-Gaussian distributions, centered at + h o and − h o , with the same width σ. This distribution is very general and can recover in appropriate limits the bimodal distribution ( ) and the single-Gaussian one ( h o = 0). We performed Monte Carlo simulations in lattices with linear sizes in the range L = 32–512. The system exhibits ferromagnetic and paramagnetic steady states. Our results suggest the occurrence of first-order phase transitions between the above-mentioned phases at low temperatures and large random-field intensities h o for some small values of the width σ. By means of finite size scaling, we estimate the critical exponents in the low-field region, where we have continuous phase transitions. In addition, we show a sketch of the phase diagram of the model for some values of σ.
We consider the transverse field Ising model with additional all-to-all interactions between the spins. We show that a mean-field treatment of this model becomes exact in the thermodynamic limit, despite the presence of 1D short-range interactions. Namely, we show that the eigenstates of the model are coherent states with an amplitude that varies through the Hilbert space, within which expectation values of local observables can be computed with mean-field theory. We study then the thermodynamics of the model and identify the different phases. Among its peculiar features, this 1D model possesses a second-order phase transition at finite temperature and exhibits inverse melting.
We consider a system of linear chains of Ising spins with short-range nearest-neighbor ferromagnetic interactions of strength K along the chains, and with infinite-range random interchain interactions of a Sherrington-Kirkpatrick type. Also included in the model is a local Gaussian random field with variance \ensuremath{\Delta}. In the replica symmetric case, the system is mapped exactly on the one-dimensional Ising model in a random field and a recursion method for calculating the spin-correlation functions is developed. For \ensuremath{\Delta}=0, an analytic relation for freezing temperature Tf{\mathit{T}}_{\mathit{f}}Tf​(K) is obtained. The replica-symmetric spin-glass phase is shown to be stable above the freezing temperature Tf{\mathit{T}}_{\mathit{f}}Tf​(K,\ensuremath{\Delta}), which is determined numerically.