arXiv subjects

Nicolas Sourlas

Publications and source records attributed to Nicolas Sourlas.

18 recordsLinked to original sources

arXiv2024

Time irreversibility in Statistical Mechanics

One of the important questions in statistical mechanics is how irreversibility (time's arrow) occurs when Newton equations of motion are time reversal invariant. One objection to irreversibility is based on Poincaré's recursion theorem: a classical hamiltonian confined system returns after some time, so-called Poincaré recurrence time (PRT), close to its initial configuration. Boltzmann's reply was that for a $N \sim 10^{23} $ macroscopic number of particles, PRT is very large and exceeds the age of the universe. In this paper we compute for the first time, using molecular dynamics, a typical recurrence time $ T(N)$ for a realistic case of a gas of $N$ particles. We find that $T(N) \sim N^z \exp (y N) $ and determine the exponents $y$ and $z$ for different values of the particle density and temperature. We also compute $y$ analytically using Boltzmann's hypotheses. We find an excellent agreement with the numerical results. This agreement validates Boltzmann's hypotheses which are not yet mathematically proven. We establish that that $T(N) $ exceeds the age of the Universe for a relatively small number of particles, much smaller than $ 10^{23} $.

cond-mat.stat-mech↗

arXiv2023

Finite-size scaling of the random-field Ising model above the upper critical dimension

Finite-size scaling above the upper critical dimension is a long-standing puzzle in the field of Statistical Physics. Even for pure systems various scaling theories have been suggested, partially corroborated by numerical simulations. In the present manuscript we address this problem in the even more complicated case of disordered systems. In particular, we investigate the scaling behavior of the random-field Ising model at dimension $D = 7$, i.e., above its upper critical dimension $D_{\rm u} = 6$, by employing extensive ground-state numerical simulations. Our results confirm the hypothesis that at dimensions $D > D_{\rm u}$, linear length scale $L$ should be replaced in finite-size scaling expressions by the effective scale $L_{\rm eff} = L^{D / D_{\rm u}}$. Via a fitted version of the quotients method that takes this modification, but also subleading scaling corrections into account, we compute the critical point of the transition for Gaussian random fields and provide estimates for the full set of critical exponents. Thus, our analysis indicates that this modified version of finite-size scaling is successful also in the context of the random-field problem.

cond-mat.stat-mech↗

arXiv2019

On the critical exponent $α$ of the 5D random-field Ising model

We present a complementary estimation of the critical exponent $α$ of the specific heat of the 5D random-field Ising model from zero-temperature numerical simulations. Our result $α= 0.12(2)$ is consistent with the estimation coming from the modified hyperscaling relation and provides additional evidence in favor of the recently proposed restoration of dimensional reduction in the random-field Ising model at $D = 5$.

cond-mat.stat-mech↗

arXiv2019

Fermionic duality

Duality transformations play a very important role in theoretical physics. In this paper I propose new duality transformations for fermionic theories. They map the strong coupling regime of one theory to the weak coupling regime of another theory. These transformations are based on the functional integral representation of the fermionic theories in terms of Grassmann variables and rely heavily on the properties of Grassmann variables. Potential applications include the study of the strong coupling phase of the two dimensional Hubbard model.

cond-mat.stat-mech↗

arXiv2019

Evidence for Supersymmetry in the Random-Field Ising Model at D = 5

We provide a non-trivial test of supersymmetry in the random-field Ising model at five spatial dimensions, by means of extensive zero-temperature numerical simulations. Indeed, supersymmetry relates correlation functions in a D-dimensional disordered system with some other correlation functions in a D-2 clean system. We first show how to check these relationships in a finite-size scaling calculation, and then perform a high-accuracy test. While the supersymmetric predictions are satisfied even to our high-accuracy at D=5, they fail to describe our results at D=4.

cond-mat.stat-mech↗

arXiv2017

Review of recent developments in the random-field Ising model

A lot of progress has been made recently in our understanding of the random-field Ising model thanks to large-scale numerical simulations. In particular, it has been shown that, contrary to previous statements: the critical exponents for different probability distributions of the random fields and for diluted antiferromagnets in a field are the same. Therefore, critical universality, which is a perturbative renormalization-group prediction, holds beyond the validity regime of perturbation theory. Most notably, dimensional reduction is restored at five dimensions, i.e., the exponents of the random-field Ising model at five dimensions and those of the pure Ising ferromagnet at three dimensions are the same.

cond-mat.dis-nn↗

arXiv2017

The $ε$ expansion and Universality in three dimensions

It has been observed that the clasification into universality classes of critical behaviour, as established by perturbative renormalization group in the viscinity of four or six dimensions of space by the epsilon expansion, remains valid down to three dimensions in all known cases, even when purturbative renormalisation group fails in three dimensions. In this paper we argue that this classification into universality classes remains true in lower dimensions of space, even when purturbative renormalisation group fails, because of the well known phenomenon of eigenvalue repulsion.

cond-mat.stat-mech↗

arXiv2017

Restoration of dimensional reduction in the random-field Ising model at five dimensions

The random-field Ising model is one of the few disordered systems where the perturbative renormalization group can be carried out to all orders of perturbation theory. This analysis predicts dimensional reduction, i.e., that the critical properties of the random-field Ising model in $D$ dimensions are identical to those of the pure Ising ferromagnet in $D-2$ dimensions. It is well known that dimensional reduction is not true in three dimensions, thus invalidating the perturbative renormalization group prediction. Here, we report high-precision numerical simulations of the 5D random-field Ising model at zero temperature. We illustrate universality by comparing different probability distributions for the random fields. We compute all the relevant critical exponents (including the critical slowing down exponent for the ground-state finding algorithm), as well as several other renormalization-group invariants. The estimated values of the critical exponents of the 5D random-field Ising model are statistically compatible to those of the pure 3D Ising ferromagnet. These results support the restoration of dimensional reduction at $D = 5$. We thus conclude that the failure of the perturbative renormalization group is a low-dimensional phenomenon. We close our contribution by comparing universal quantities for the random-field problem at dimensions $3 \leq D < 6$ to their values in the pure Ising model at $D-2$ dimensions and we provide a clear verification of the Rushbrooke equality at all studied dimensions.

cond-mat.dis-nn↗

arXiv2016

Phase transitions in disordered systems: the example of the random-field Ising model in four dimensions

By performing a high-statistics simulation of the $D=4$ random-field Ising model at zero temperature for different shapes of the random-field distribution, we show that the model is ruled by a single universality class. We compute to a high accuracy the complete set of critical exponents for this class, including the correction-to-scaling exponent. Our results indicate that in four dimensions: (i) dimensional reduction as predicted by the perturbative renormalization group does not hold and (ii) three independent critical exponents are needed to described the transition.

cond-mat.dis-nn↗

arXiv2014

Diluted Antiferromagnetic 3D Ising model in a field

We present numerical simulations for the diluted antiferromagnetic 3D Ising model (DAFF) in an external magnetic field at zero temperature. Our results are compatible with the DAFF being in the same universality class as the Random Field Ising model, in agreement with the renormalization group prediction.

cond-mat.dis-nn↗

arXiv2014

On the phase transition of the 3D random field Ising model

We present numerical simulations of the random field Ising model in three dimensions at zero temperature. The critical exponents are found to agree with previous results. We study the magnetic susceptibility by applying a small magnetic field perturbation. We find that the critical amplitude ratio of the magnetic susceptibilities to be very large, equal to 233.1 \pm 1.5. We find strong sample to sample fluctuations which obey finite size scaling. The probability distribution of the size of small energy excitations is maximally non-self averaging, obeying a double peak distribution, and is finite size scaling invariant. We also study the approach to the thermodynamic limit of the ground state magnetization at the phase transition.

cond-mat.dis-nn↗

arXiv2004

Soft annealing: A new approach to difficult computational problems

I propose a new method to study computationally difficult problems. I consider a new system, larger than the one I want to simulate. The original system is recovered by imposing constraints on the large system. I simulate the large system with the hard constraints replaced by soft constraints. I illustrate the method in the case of the ferromagnetic Ising model and in the case the three dimensional spin-glass model. I show that in both models the phases of the soft problem have the same properties as the phases of the original model and that the softened model belongs to the same universality class as the original one. I show that correlation times are much shorter in the larger soft constrained system and that it is computationally advantageous to study it instead of the original system. This method is quite general and can be applied to many other systems.

cond-mat.dis-nn↗

arXiv2003

Scale Invariance and Self-averaging in disordered systems

In a previous paper we found that in the random field Ising model at zero temperature in three dimensions the correlation length is not self-averaging near the critical point and that the violation of self-averaging is maximal. This is due to the formation of bound states in the underlying field theory. We present a similar study for the case of disordered Potts and Ising ferromagnets in two dimensions near the critical temperature. In the random Potts model the correlation length is not self-averaging near the critical temperature but the violation of self-averaging is weaker than in the random field case. In the random Ising model we find still weaker violations of self-averaging and we cannot rule out the possibility of the restoration of self-averaging in the infinite volume limit.

cond-mat.stat-mech↗

arXiv2002

Scale Invariance in disordered systems: the example of the Random Field Ising Model

We show by numerical simulations that the correlation function of the random field Ising model (RFIM) in the critical region in three dimensions has very strong fluctuations and that in a finite volume the correlation length is not self-averaging. This is due to the formation of a bound state in the underlying field theory. We argue that this non perturbative phenomenon is not particular to the RFIM in 3-d. It is generic for disordered systems in two dimensions and may also happen in other three dimensional disordered systems.

cond-mat↗

arXiv2001

Statistical Mechanics and Capacity-Approaching Error-Correcting Codes

I will show that there is a deep relation between error-correction codes and certain mathematical models of spin glasses. In particular minimum error probability decoding is equivalent to finding the ground state of the corresponding spin system. The most probable value of a symbol is related to the magnetization at a different temperature. Convolutional codes correspond to one-dimensional spin systems and Viterbi's decoding algorithm to the transfer matrix algorithm of Statistical Mechanics. I will also show how the recently discovered (or rediscovered) capacity approaching codes (turbo codes and low density parity check codes) can be analysed using statistical mechanics. It is possible to show, using statistical mechanics, that these codes allow error-free communication for signal to noise ratio above a certain threshold. This threshold depends on the particular code, and can be computed analytically in many cases.

cond-mat↗

arXiv1999

P-adic numbers and replica symmetry breaking

The p-adic formulation of replica symmetry breaking is presented. In this approach ultrametricity is a natural consequence of the basic properties of the p-adic numbers. Many properties can be simply derived in this approach and p-adic Fourier transform seems to be an promising tool.

cond-mat.dis-nn↗

arXiv1998

Statistical Mechanics and error-correction Codes

I will show that there is a deep relation between error-correction codes and certain mathematical models of spin glasses. In particular minimum error probability decoding is equivalent to finding the ground state of the corresponding spin system. The most probable value of a symbol is related to the magnetization at a different temperature. Convolutional codes correspond to one-dimensional spin systems and Viterbi's decoding algorithm to the transfer matrix algorithm of Statistical Mechanics. A particular spin-glass model, which is exactly soluble, corresponds to an ideal code, i.e. a code which allows error-free communication if the rate is below channel capacity.

cond-mat.stat-mech↗

arXiv1998

Universality in Random Systems: the case of the 3-d Random Field Ising model

We study numerically the zero temperature Random Field Ising Model on cubic lattices of various linear sizes $ 6 \le L \le 90 $ in three dimensions with the purpose of verifying the validity of universality for disordered systems. For each random field configuration we vary the ferromagnetic coupling strength J and compute the ground state exactly. We examine the case of different random field probability distributions: gaussian distribution, zero width bimodal distribution h_{i} = \pm 1, wide bimodal distribution h_{i} = \pm 1 +δh (with a gaussian $δh$). We also study the case of the randomly diluted antiferromagnet in a field,which is thought to be in the same universality class. We find that in the infinite volume limit the magnetization is discontinuous in J and we compute the relevant exponent, which, according to finite size scaling, equals $ 1/ ν$ . We find different values of $ ν$ for the different random field distributions, in disagreement with universality.

cond-mat.stat-mech↗