Narrow Results

In this paper we compare the relaxation in several versions of the Sznajd model (SM) with random sequential updating on the chain and square lattice. We start by reviewing briefly all proposed one-dimensional versions of SM. Next, we compare the results obtained from Monte Carlo simulations with the mean field results obtained by Slanina and Lavicka. Finally, we investigate the relaxation on the square lattice and compare two generalizations of SM, one suggested by Stauffer et al. and another by Galam. We show that there are no qualitative differences between these two approaches, although the relaxation within the Galam rule is faster than within the well known Stauffer et al. rule.

We have studied the magnetic relaxation behavior of a two-dimensional Ising ferromagnet by Monte Carlo simulation. Our primary goal is to investigate the effects of the system’s geometry (area preserving), boundary conditions and dynamical rules on the relaxation behavior. The Glauber and Metropolis dynamical rules have been employed. The systems with periodic and open boundary conditions are studied. The major findings are the exponential relaxation and the dependence of relaxation time ($\tau$) on the aspect ratio $R$ (length over breadth having fixed area). A power law dependence ($\tau \sim R^{-s}$) has been observed for larger values of aspect ratio ($R$). The exponent ($s$) has been found to depend linearly ($s=aT+b$) on the system’s temperature ($T$). The transient behaviors of the spin-flip density have been investigated for both surface and bulk/core. The size dependencies of saturated spin-flip density significantly differ for the surface and the bulk/core. Both the saturated bulk/core and saturated surface spin-flip density was found to follow the logarithmic dependence $f_d=a+blog(L)$ with the system size. The faster relaxation was observed for open boundary condition with any kind (Metropolis/Glauber) of dynamical rule. Similarly, Metropolis algorithm yields faster relaxation for any kind (open/periodic) of boundary condition.

The extension of dynamical scaling to local, space-time dependent rescaling factors is investigated. For a dynamical exponent z=2, the corresponding invariance group is the Schrödinger group. Schrödinger invariance is shown to determine completely the two-point correlation function. The result is checked in two exactly solvable models.

The damage spreading of the Ising model on a set of trivalent structures is studied with competing Glauber and Kawasaki dynamics. We find that the damage transition temperature T_d increases as the number of rectangles increases at the pure Glauber dynamics case and T_d decreases sharply as the probability of occurrence of Kawasaki dynamics increases. A heuristic argument is provided to explain the numerical results.

We construct a new equilibrium dynamics of infinite particle systems in a Riemannian manifold X. This dynamics is an analog of the Kawasaki dynamics of lattice spin systems. The Kawasaki dynamics now is a process where interacting particles randomly hop over X. We establish conditions on the a priori explicitly given symmetrizing measure and the generator of this dynamics, under which a corresponding conservative Markov processes exists. We also outline two types of scaling limit of the equilibrium Kawasaki dynamics: one leading to an equilibrium Glauber dynamics in continuum (a birth-and-death process), and the other leading to a diffusion dynamics of interacting particles (in particular, the gradient stochastic dynamics).

We consider Vlasov-type scaling for the Glauber dynamics in continuum with a positive integrable potential, and construct rescaled and limiting evolutions of correlation functions. Convergence to the limiting evolution for the positive density system in infinite volume is shown. Chaos preservation property of this evolution gives a possibility to derive a nonlinear Vlasov-type equation for the particle density of the limiting system.

The nature of the dynamics of opinion formation or zero-temperature Ising models modeled as a decision-by-majority process in complex networks is investigated using eigenmode analysis. The Hamiltonian of the system is defined and estimated by eigenvectors of the adjacency matrix constructed from several network models. The rule of the process is assumed to be equivalent to the minimization of the Hamiltonian. The initial and final states of the dynamics are decomposed on the basis of the eigenvectors. The process and the eigenmodes are analyzed by numerical studies. We show that the magnitude of the coefficient for the largest eigenvector at the initial states is the key determinant for the resulting dynamics. We thus prove that the final state of the dynamics can be estimated by the eigenmodes of the initial state.