Showing 1–47 of 47 results for author: Berg, B

Warming or cooling from a random walk process in the temperature

Authors: Bernd Albert Berg

Abstract: A simple 3-parameter random walk model for monthly fluctuations $\triangle T$ of a temperature $T$ is introduced. Applied to a time range of 170 years, temperature fluctuations of the model produce for about 14\% of the runs warming that exceeds the observed global warming of the earth surface temperature from 1850 to 2019. On the other hand, there is a 50\% likelihood for runs of our model result… ▽ More A simple 3-parameter random walk model for monthly fluctuations $\triangle T$ of a temperature $T$ is introduced. Applied to a time range of 170 years, temperature fluctuations of the model produce for about 14\% of the runs warming that exceeds the observed global warming of the earth surface temperature from 1850 to 2019. On the other hand, there is a 50\% likelihood for runs of our model resulting in cooling. If a similar random walk process can be used as an effective model for fluctuations of the global earth surface temperature, effects due to internal and external forcing could be considerably over- or underestimated. △ Less

Submitted 1 February, 2020; originally announced February 2020.

Comments: 4 pages, 8 figures

A brief history of the introduction of generalized ensembles to Markov chain Monte Carlo simulations

Authors: Bernd A Berg

Abstract: The most efficient weights for Markov chain Monte Carlo calculations of physical observables are not necessarily those of the canonical ensemble. Generalized ensembles, which do not exist in nature but can be simulated on computers, lead often to a much faster convergence. In particular, they have been used for simulations of first order phase transitions and for simulations of complex systems in… ▽ More The most efficient weights for Markov chain Monte Carlo calculations of physical observables are not necessarily those of the canonical ensemble. Generalized ensembles, which do not exist in nature but can be simulated on computers, lead often to a much faster convergence. In particular, they have been used for simulations of first order phase transitions and for simulations of complex systems in which conflicting constraints lead to a rugged free energy landscape. Starting off with the Metropolis algorithm and Hastings' extension, I present a mini review which focuses on the explosive use of generalized ensembles in the early 1990s. Illustrations are given, which range from spin models to peptides. △ Less

Submitted 13 December, 2016; originally announced December 2016.

Comments: 15 pages, 9 figure, based on an invited talk given at the Conference on Phase Transitions and Critical Phenomena, Coventry, April 6-8, 2016

Least square fitting with one parameter less

Authors: Bernd A. Berg

Abstract: It is shown that whenever the multiplicative normalization of a fitting function is not known, least square fitting by $χ^2$ minimization can be performed with one parameter less than usual by converting the normalization parameter into a function of the remaining parameters and the data. It is shown that whenever the multiplicative normalization of a fitting function is not known, least square fitting by $χ^2$ minimization can be performed with one parameter less than usual by converting the normalization parameter into a function of the remaining parameters and the data. △ Less

Submitted 2 May, 2016; v1 submitted 28 May, 2015; originally announced May 2015.

Comments: 6 pages, 1 figure. Fortran code available on the Web. Erratum: The 4-parameter example suffered from a typo in two subroutines, which is now corrected

Journal ref: Comp. Phys. Comm. 200 (2016) 254-258

Density of states and Fisher's zeros in compact U(1) pure gauge theory

Authors: A. Bazavov, B. A. Berg, Daping Du, Y. Meurice

Abstract: We present high-accuracy calculations of the density of states using multicanonical methods for lattice gauge theory with a compact gauge group U(1) on 4^4, 6^4 and 8^4 lattices. We show that the results are consistent with weak and strong coupling expansions. We present methods based on Chebyshev interpolations and Cauchy theorem to find the (Fisher's) zeros of the partition function in the compl… ▽ More We present high-accuracy calculations of the density of states using multicanonical methods for lattice gauge theory with a compact gauge group U(1) on 4^4, 6^4 and 8^4 lattices. We show that the results are consistent with weak and strong coupling expansions. We present methods based on Chebyshev interpolations and Cauchy theorem to find the (Fisher's) zeros of the partition function in the complex beta=1/g^2 plane. The results are consistent with reweighting methods whenever the latter are accurate. We discuss the volume dependence of the imaginary part of the Fisher's zeros, the width and depth of the plaquette distribution at the value of beta where the two peaks have equal height. We discuss strategies to discriminate between first and second order transitions and explore them with data at larger volume but lower statistics. Higher statistics and even larger lattices are necessary to draw strong conclusions regarding the order of the transition. △ Less

Submitted 9 February, 2012; originally announced February 2012.

Comments: 14 pages, 16 figures

Report number: BNL-96917-2012-JA

Journal ref: Phys. Rev. D 85, 056010 (2012)

Display of probability densities for data from a continuous distribution

Authors: Bernd A. Berg

Abstract: Based on cumulative distribution functions, Fourier series expansion and Kolmogorov tests, we present a simple method to display probability densities for data drawn from a continuous distribution. It is often more efficient than using histograms. Based on cumulative distribution functions, Fourier series expansion and Kolmogorov tests, we present a simple method to display probability densities for data drawn from a continuous distribution. It is often more efficient than using histograms. △ Less

Submitted 3 May, 2011; originally announced May 2011.

Comments: 5 pages, 4 figures, presented at Computer Simulation Studies XXIV, Athens, GA, 2011

Journal ref: Physics Procedia 15 (2011) 17-21

Commuting Heisenberg operators as the quantum response problem: Time-normal averages in the truncated Wigner representation

Authors: B. Berg, L. I. Plimak, A. Polkovnikov, M. K. Olsen, M. Fleischhauer, W. P. Schleich

Abstract: The applicability of the so-called truncated Wigner approximation (-W) is extended to multitime averages of Heisenberg field operators. This task splits naturally in two. Firstly, what class of multitime averages the -W approximates, and, secondly, how to proceed if the average in question does not belong to this class. To answer the first question we develop an (in principle, exact) path-integr… ▽ More The applicability of the so-called truncated Wigner approximation (-W) is extended to multitime averages of Heisenberg field operators. This task splits naturally in two. Firstly, what class of multitime averages the -W approximates, and, secondly, how to proceed if the average in question does not belong to this class. To answer the first question we develop an (in principle, exact) path-integral approach in phase-space based on the symmetric (Weyl) ordering of creation and annihilation operators. These techniques calculate a new class of averages which we call time-symmetric. The -W equations emerge as an approximation within this path-integral techniques. We then show that the answer to the second question is associated with response properties of the system. In fact, for two-time averages Kubo's renowned formula relating the linear response function to two-time commutators suffices. The -W is trivially generalised to the response properties of the system allowing one to calculate approximate time-normally ordered two-time correlation functions with surprising ease. The techniques we develop are demonstrated for the Bose-Hubbard model. △ Less

Submitted 19 May, 2009; originally announced May 2009.

Comments: 20 pages, 6 figures

Journal ref: Phys. Rev. A 80, 033624 (2009)

Program package for multicanonical simulations of U(1) lattice gauge theory

Authors: Alexei Bazavov, Bernd A. Berg

Abstract: We document our Fortran 77 code for multicanonical simulations of 4D U(1) lattice gauge theory in the neighborhood of its phase transition. This includes programs and routines for canonical simulations using biased Metropolis heatbath updating and overrelaxation, determination of multicanonical weights via a Wang-Landau recursion, and multicanonical simulations with fixed weights supplemented by… ▽ More We document our Fortran 77 code for multicanonical simulations of 4D U(1) lattice gauge theory in the neighborhood of its phase transition. This includes programs and routines for canonical simulations using biased Metropolis heatbath updating and overrelaxation, determination of multicanonical weights via a Wang-Landau recursion, and multicanonical simulations with fixed weights supplemented by overrelaxation sweeps. Measurements are performed for the action, Polyakov loops and some of their structure factors. Many features of the code transcend the particular application and are expected to be useful for other lattice gauge theory models as well as for systems in statistical physics. △ Less

Submitted 23 March, 2009; originally announced March 2009.

Comments: 10 pages, 7 figures

Journal ref: Comput.Phys.Commun.180:2339-2347,2009

Thermodynamics of two lattice ice models in three dimensions

Authors: Chizuru Muguruma, Yuko Okamoto, Bernd A. Berg

Abstract: In a recent paper we introduced two Potts-like models in three dimensions, which share the following properties: (A) One of the ice rules is always fulfilled (in particular also at infinite temperature). (B) Both ice rules hold for groundstate configurations. This allowed for an efficient calculation of the residual entropy of ice I (ordinary ice) by means of multicanonical simulations. Here we… ▽ More In a recent paper we introduced two Potts-like models in three dimensions, which share the following properties: (A) One of the ice rules is always fulfilled (in particular also at infinite temperature). (B) Both ice rules hold for groundstate configurations. This allowed for an efficient calculation of the residual entropy of ice I (ordinary ice) by means of multicanonical simulations. Here we present the thermodynamics of these models. Despite their similarities with Potts models, no sign of a disorder-order phase transition is found. △ Less

Submitted 3 June, 2008; originally announced June 2008.

Comments: 5 pages, 7 figures

Journal ref: Phys. Rev. E 78 (2008) 041113

Phase Transition Properties of 3D Potts Models

Authors: Alexei Bazavov, Bernd A. Berg, Santosh Dubey

Abstract: Using multicanonical Metropolis simulations we estimate phase transition properties of 3D Potts models for q=4 to 10: The transition temperatures, latent heats, entropy gaps, normalized entropies at the disordered and ordered endpoints, interfacial tensions, and spinodal endpoints. Using multicanonical Metropolis simulations we estimate phase transition properties of 3D Potts models for q=4 to 10: The transition temperatures, latent heats, entropy gaps, normalized entropies at the disordered and ordered endpoints, interfacial tensions, and spinodal endpoints. △ Less

Submitted 12 July, 2008; v1 submitted 8 April, 2008; originally announced April 2008.

Comments: 18 pages, 7 figures. Correction of minor typos, one reference added

Journal ref: Nucl.Phys.B802:421-434,2008

Finite Volume Kolmogorov-Johnson-Mehl-Avrami Theory

Authors: Bernd A. Berg, Santosh Dubey

Abstract: We study Kolmogorov-Johnson-Mehl-Avrami (KJMA) theory of phase conversion in finite volumes. For the conversion time we find the relationship $τ_{\rm con} = τ_{\rm nu} [1+f_d(q)]$. Here $d$ is the space dimension, $τ_{\rm nu}$ the nucleation time in the volume $V$, and $f_d(q)$ a scaling function. Its dimensionless argument is $q=τ_{\rm ex}/ τ_{\rm nu}$, where $τ_{\rm ex}$ is an expansion time,… ▽ More We study Kolmogorov-Johnson-Mehl-Avrami (KJMA) theory of phase conversion in finite volumes. For the conversion time we find the relationship $τ_{\rm con} = τ_{\rm nu} [1+f_d(q)]$. Here $d$ is the space dimension, $τ_{\rm nu}$ the nucleation time in the volume $V$, and $f_d(q)$ a scaling function. Its dimensionless argument is $q=τ_{\rm ex}/ τ_{\rm nu}$, where $τ_{\rm ex}$ is an expansion time, defined to be proportional to the diameter of the volume divided by expansion speed. We calculate $f_d(q)$ in one, two and three dimensions. The often considered limits of phase conversion via either nucleation or spinodal decomposition are found to be volume-size dependent concepts, governed by simple power laws for $f_d(q)$. △ Less

Submitted 22 April, 2008; v1 submitted 4 February, 2008; originally announced February 2008.

Comments: 4 pages, 4 figures. Additions after referee reports: Scaling of the variable q is proven. Additional references are added

Journal ref: Phys.Rev.Lett.100:165702,2008

From Data to Probability Densities without Histograms

Authors: Bernd A. Berg, Robert C. Harris

Abstract: When one deals with data drawn from continuous variables, a histogram is often inadequate to display their probability density. It deals inefficiently with statistical noise, and binsizes are free parameters. In contrast to that, the empirical cumulative distribution function (obtained after sorting the data) is parameter free. But it is a step function, so that its differentiation does not give… ▽ More When one deals with data drawn from continuous variables, a histogram is often inadequate to display their probability density. It deals inefficiently with statistical noise, and binsizes are free parameters. In contrast to that, the empirical cumulative distribution function (obtained after sorting the data) is parameter free. But it is a step function, so that its differentiation does not give a smooth probability density. Based on Fourier series expansion and Kolmogorov tests, we introduce a simple method, which overcomes this problem. Error bars on the estimated probability density are calculated using a jackknife method. We give several examples and provide computer code reproducing them. You may want to look at the corresponding figures 4 to 9 first. △ Less

Submitted 22 December, 2007; originally announced December 2007.

Comments: 9 pages, 9 figures

Journal ref: Comp. Phys. Commun. 179 (2008) 443-448

Markov-Chain Monte Carlo Methods for Simulations of Biomolecules

Authors: Bernd A. Berg

Abstract: The computer revolution has been driven by a sustained increase of computational speed of approximately one order of magnitude (a factor of ten) every five years since about 1950. In natural sciences this has led to a continuous increase of the importance of computer simulations. Major enabling techniques are Markov Chain Monte Carlo (MCMC) and Molecular Dynamics (MD) simulations. This article… ▽ More The computer revolution has been driven by a sustained increase of computational speed of approximately one order of magnitude (a factor of ten) every five years since about 1950. In natural sciences this has led to a continuous increase of the importance of computer simulations. Major enabling techniques are Markov Chain Monte Carlo (MCMC) and Molecular Dynamics (MD) simulations. This article deals with the MCMC approach. First basic simulation techniques, as well as methods for their statistical analysis are reviewed. Afterwards the focus is on generalized ensembles and biased updating, two advanced techniques, which are of relevance for simulations of biomolecules, or are expected to become relevant with that respect. In particular we consider the multicanonical ensemble and the replica exchange method (also known as parallel tempering or method of multiple Markov chains). △ Less

Submitted 4 September, 2007; originally announced September 2007.

Comments: 34 pages, 13 figures, review article

Journal ref: Springer Lecture Notes in Physics 736, 319-353 (2008)

Numerical calculation of the combinatorial entropy of partially ordered ice

Authors: Bernd A. Berg, Wei Yang

Abstract: Using a one-parameter case as an example, we demonstrate that multicanonical simulations allow for accurate estimates of the residual combinatorial entropy of partially ordered ice. For the considered case corrections to an (approximate) analytical formula are found to be small, never exceeding 0.5%. The method allows one as well to calculate combinatorial entropies for many other systems. Using a one-parameter case as an example, we demonstrate that multicanonical simulations allow for accurate estimates of the residual combinatorial entropy of partially ordered ice. For the considered case corrections to an (approximate) analytical formula are found to be small, never exceeding 0.5%. The method allows one as well to calculate combinatorial entropies for many other systems. △ Less

Submitted 20 September, 2007; v1 submitted 11 July, 2007; originally announced July 2007.

Comments: Extended version: 7 pages, 10 figures (v1 is letter-type version)

Journal ref: J. Chem. Phys. 127, 224502 (2007)

Normalized entropy density of the 3D 3-state Potts model

Authors: Alexei Bazavov, Bernd A. Berg

Abstract: Using a multicanonical Metropolis algorithm we have performed Monte Carlo simulations of the 3D 3-state Potts model on $L^3$ lattices with L=20, 30, 40, 50. Covering a range of inverse temperatures from $β_{\min}=0$ to $β_{\max}=0.33$ we calculated the infinite volume limit of the entropy density $s(β)$ with its normalization obtained from $s(0)=\ln 3$. At the transition temperature the entropy… ▽ More Using a multicanonical Metropolis algorithm we have performed Monte Carlo simulations of the 3D 3-state Potts model on $L^3$ lattices with L=20, 30, 40, 50. Covering a range of inverse temperatures from $β_{\min}=0$ to $β_{\max}=0.33$ we calculated the infinite volume limit of the entropy density $s(β)$ with its normalization obtained from $s(0)=\ln 3$. At the transition temperature the entropy and energy endpoints in the ordered and disordered phase are estimated employing a novel reweighting procedure. We also evaluate the transition temperature and the order-disorder interface tension. The latter estimate increases when capillary waves are taken into account. △ Less

Submitted 17 February, 2007; originally announced February 2007.

Comments: 5 pages, 4 figures

Journal ref: Phys.Rev.D75:094506,2007

Wang-Landau/Multibondic Cluster Simulations for Second-Order Phase Transitions

Authors: Bernd A. Berg, Wolfhard Janke

Abstract: For a second-order phase transition the critical energy range of interest is larger than the energy range covered by a canonical Monte Carlo simulation at the critical temperature. Such an extended energy range can be covered by performing a Wang-Landau recursion for the spectral density followed by a multicanonical simulation with fixed weights. But in the conventional approach one loses the ad… ▽ More For a second-order phase transition the critical energy range of interest is larger than the energy range covered by a canonical Monte Carlo simulation at the critical temperature. Such an extended energy range can be covered by performing a Wang-Landau recursion for the spectral density followed by a multicanonical simulation with fixed weights. But in the conventional approach one loses the advantage due to cluster algorithms. A cluster version of the Wang-Landau recursion together with a subsequent multibondic simulation improves for 2D and 3D Ising models the efficiency of the conventional Wang-Landau/multicanonical approach by power laws in the lattice size. In our simulations real gains in CPU time reach two orders of magnitude. △ Less

Submitted 11 December, 2006; v1 submitted 23 October, 2006; originally announced October 2006.

Comments: 4 pages, 3 figures, RevTex style, updated version, to appear in Phys. Rev. Lett. (in print)

Journal ref: Phys.Rev.Lett.98:040602,2007

Critical Exponents for U(1) Lattice Gauge Theory at Finite Temperature

Authors: Bernd A. Berg, Alexei Bazavov

Abstract: For compact U(1) lattice gauge theory (LGT) we have performed a finite size scaling analysis on $N_τ N_s^3$ lattices for $N_τ$ fixed and $N_s\to\infty$, approaching the phase transition from the confined phase. For $N_τ=4$, 5 and 6 our data contradict the expected scenario that this transition is either first order or in the universality class of the 3d XY model. If there are no conceptional fla… ▽ More For compact U(1) lattice gauge theory (LGT) we have performed a finite size scaling analysis on $N_τ N_s^3$ lattices for $N_τ$ fixed and $N_s\to\infty$, approaching the phase transition from the confined phase. For $N_τ=4$, 5 and 6 our data contradict the expected scenario that this transition is either first order or in the universality class of the 3d XY model. If there are no conceptional flaws in applying the argument that the Gaussian fixed point in 3d is unstable to our systems, estimates of the critical exponents $α/ν$, $γ/ν$, $(1-β)/ν$ and $2-η$ indicate the existence of a new, non-trivial renormalization group fixed point for second order phase transitions in 3d. Such a fixed point would be of importance for renormalization group theory and statistical physics. △ Less

Submitted 6 September, 2006; originally announced September 2006.

Comments: Lattice Conference, July 23-28, 2006 Tucson, Arizona, USA. 7 pages, 3 figures

Journal ref: PoSLAT2006:061,2006

Residual Entropy of Ordinary Ice from Multicanonical Simulations

Authors: Bernd A. Berg, Chizuru Muguruma, Yuko Okamoto

Abstract: We introduce two simple models with nearest neighbor interactions on 3D hexagonal lattices. Each model allows one to calculate the residual entropy of ice I (ordinary ice) by means of multicanonical simulations. This gives the correction to the residual entropy derived by Linus Pauling in 1935. Our estimate is found to be within less than 0.1% of an analytical approximation by Nagle, which is an… ▽ More We introduce two simple models with nearest neighbor interactions on 3D hexagonal lattices. Each model allows one to calculate the residual entropy of ice I (ordinary ice) by means of multicanonical simulations. This gives the correction to the residual entropy derived by Linus Pauling in 1935. Our estimate is found to be within less than 0.1% of an analytical approximation by Nagle, which is an improvement of Pauling's result. We pose it as a challenge to experimentalists to improve on the accuracy of a 1936 measurement by Giauque and Stout by about one order of magnitude, which would allow one to identify corrections to Pauling's value unambiguously. It is straightforward to transfer our methods to other crystal systems. △ Less

Submitted 19 January, 2007; v1 submitted 8 September, 2006; originally announced September 2006.

Comments: 3 pages, 2 figures. One reference added and changes after comments by John Nagle

Journal ref: Phys. Rev. B 75, 092202 (2007)

Non-Perturbative U(1) Gauge Theory at Finite Temperature

Authors: Bernd A. Berg, Alexei Bazavov

Abstract: For compact U(1) lattice gauge theory (LGT) we have performed a finite size scaling analysis on $N_τ N_s^3$ lattices for $N_τ$ fixed by extrapolating spatial volumes of size $N_s\le 18$ to $N_s\to\infty$. Within the numerical accuracy of the thus obtained fits we find for $N_τ=4$, 5 and~6 second order critical exponents, which exhibit no obvious $N_τ$ dependence. The exponents are consistent wit… ▽ More For compact U(1) lattice gauge theory (LGT) we have performed a finite size scaling analysis on $N_τ N_s^3$ lattices for $N_τ$ fixed by extrapolating spatial volumes of size $N_s\le 18$ to $N_s\to\infty$. Within the numerical accuracy of the thus obtained fits we find for $N_τ=4$, 5 and~6 second order critical exponents, which exhibit no obvious $N_τ$ dependence. The exponents are consistent with 3d Gaussian values, but not with either first order transitions or the universality class of the 3d XY model. As the 3d Gaussian fixed point is known to be unstable, the scenario of a yet unidentified non-trivial fixed point close to the 3d Gaussian emerges as one of the possible explanations. △ Less

Submitted 15 October, 2006; v1 submitted 29 May, 2006; originally announced May 2006.

Comments: Extended version after referee reports. 6 pages, 6 figures

Journal ref: Phys.Rev.D74:094502,2006

Heat Bath Efficiency with Metropolis-Type Updating

Authors: Alexei Bazavov, Bernd A. Berg

Abstract: We illustrate for 4D SU(2) and U(1) lattice gauge theory that sampling with a biased Metropolis scheme is essentially equivalent to using the heat bath algorithm. Only, the biased Metropolis method can also be applied when an efficient heat bath algorithm does not exist. We illustrate for 4D SU(2) and U(1) lattice gauge theory that sampling with a biased Metropolis scheme is essentially equivalent to using the heat bath algorithm. Only, the biased Metropolis method can also be applied when an efficient heat bath algorithm does not exist. △ Less

Submitted 26 April, 2005; v1 submitted 7 March, 2005; originally announced March 2005.

Comments: 5 pages, 4 figures. Revisions after referee reports

Journal ref: Phys.Rev. D71 (2005) 114506

Rugged Metropolis Sampling with Simultaneous Updating of Two Dynamical Variables

Authors: Bernd A. Berg, Huan-Xiang Zhou

Abstract: The Rugged Metropolis (RM) algorithm is a biased updating scheme, which aims at directly hitting the most likely configurations in a rugged free energy landscape. Details of the one-variable (RM$_1$) implementation of this algorithm are presented. This is followed by an extension to simultaneous updating of two dynamical variables (RM$_2$). In a test with Met-Enkephalin in vacuum RM$_2$ improves… ▽ More The Rugged Metropolis (RM) algorithm is a biased updating scheme, which aims at directly hitting the most likely configurations in a rugged free energy landscape. Details of the one-variable (RM$_1$) implementation of this algorithm are presented. This is followed by an extension to simultaneous updating of two dynamical variables (RM$_2$). In a test with Met-Enkephalin in vacuum RM$_2$ improves conventional Metropolis simulations by a factor of about four. Correlations between three or more dihedral angles appear to prevent larger improvements at low temperatures. We also investigate a multi-hit Metropolis scheme, which spends more CPU time on variables with large autocorrelation times. △ Less

Submitted 8 May, 2005; v1 submitted 3 February, 2005; originally announced February 2005.

Comments: 8 pages, 5 figures. Revisions after referee reports. Additional simulations for temperatures down to 220K

Journal ref: Phys. Rev. E 72, 016712 (2005)

Introduction to Markov Chain Monte Carlo Simulations and their Statistical Analysis

Authors: Bernd A. Berg

Abstract: This article is a tutorial on Markov chain Monte Carlo simulations and their statistical analysis. The theoretical concepts are illustrated through many numerical assignments from the author's book on the subject. Computer code (in Fortran) is available for all subjects covered and can be downloaded from the web. This article is a tutorial on Markov chain Monte Carlo simulations and their statistical analysis. The theoretical concepts are illustrated through many numerical assignments from the author's book on the subject. Computer code (in Fortran) is available for all subjects covered and can be downloaded from the web. △ Less

Submitted 19 October, 2004; originally announced October 2004.

Comments: Invited lectures given at the National University of Singapore

Journal ref: In Markov Chain Monte Carlo, W.S. Kendall et al. Editors, Lecture Notes Series, Institute for Mathematical Sciences, National University of Singapore, Vol.7, p.1 ff., World Scientific, 2005

Monte Carlo Protein Folding: Simulations of Met-Enkephalin with Solvent-Accessible Area Parameterizations

Authors: Hsiao-Ping Hsu, Bernd A. Berg, Peter Grassberger

Abstract: Treating realistically the ambient water is one of the main difficulties in applying Monte Carlo methods to protein folding. The solvent-accessible area method, a popular method for treating water implicitly, is investigated by means of Metropolis simulations of the brain peptide Met-Enkephalin. For the phenomenological energy function ECEPP/2 nine atomic solvation parameter (ASP) sets are studi… ▽ More Treating realistically the ambient water is one of the main difficulties in applying Monte Carlo methods to protein folding. The solvent-accessible area method, a popular method for treating water implicitly, is investigated by means of Metropolis simulations of the brain peptide Met-Enkephalin. For the phenomenological energy function ECEPP/2 nine atomic solvation parameter (ASP) sets are studied that had been proposed by previous authors. The simulations are compared with each other, with simulations with a distance dependent electrostatic permittivity $ε(r)$, and with vacuum simulations ($ε=2$). Parallel tempering and a recently proposed biased Metropolis technique are employed and their performances are evaluated. The measured observables include energy and dihedral probability densities (pds), integrated autocorrelation times, and acceptance rates. Two of the ASP sets turn out to be unsuitable for these simulations. For all other sets, selected configurations are minimized in search of the global energy minima. Unique minima are found for the vacuum and the $ε(r)$ system, but for none of the ASP models. Other observables show a remarkable dependence on the ASPs. In particular, autocorrelation times vary dramatically with the ASP parameters. Three ASP sets have much smaller autocorrelations at 300 K than the vacuum simulations, opening the possibility that simulations can be speeded up vastly by judiciously chosing details of the force △ Less

Submitted 26 August, 2004; originally announced August 2004.

Comments: 10 pages; published in "NIC Symposium 2004", eds. D. Wolf at el. (NIC, Juelich, 2004)

Dynamics of Phase Transitions: The 3D 3-state Potts model

Authors: Bernd A. Berg, Hildegard Meyer-Ortmanns, Alexander Velytsky

Abstract: In studies of the QCD deconfining phase transition or cross-over by means of heavy ion experiments, one ought to be concerned about non-equilibrium effects due to heating and cooling of the system. In this paper we extend our previous study of Glauber dynamics of 2D Potts models to the 3D 3-state Potts model, which serves as an effective model for some QCD properties. We investigate the linear t… ▽ More In studies of the QCD deconfining phase transition or cross-over by means of heavy ion experiments, one ought to be concerned about non-equilibrium effects due to heating and cooling of the system. In this paper we extend our previous study of Glauber dynamics of 2D Potts models to the 3D 3-state Potts model, which serves as an effective model for some QCD properties. We investigate the linear theory of spinodal decomposition in some detail. It describes the early time evolution of the 3D model under a quench from the disordered into the ordered phase well, but fails in 2D. Further, the quench leads to competing vacuum domains, which are difficult to equilibrate, even in the presence of a small external magnetic field. From our hysteresis study we find, as before, a dynamics dominated by spinodal decomposition. There is evidence that some effects survive in the case of a cross-over. But the infinite volume extrapolation is difficult to control, even with lattices as large as $120^3$. △ Less

Submitted 19 August, 2004; v1 submitted 13 May, 2004; originally announced May 2004.

Comments: 12 pages; added references, corrected typos

Journal ref: Phys.Rev. D70 (2004) 054505

Biased Metropolis Sampling for Rugged Free Energy Landscapes

Authors: Bernd A. Berg

Abstract: Metropolis simulations of all-atom models of peptides (i.e. small proteins) are considered. Inspired by the funnel picture of Bryngelson and Wolyness, a transformation of the updating probabilities of the dihedral angles is defined, which uses probability densities from a higher temperature to improve the algorithmic performance at a lower temperature. The method is suitable for canonical as wel… ▽ More Metropolis simulations of all-atom models of peptides (i.e. small proteins) are considered. Inspired by the funnel picture of Bryngelson and Wolyness, a transformation of the updating probabilities of the dihedral angles is defined, which uses probability densities from a higher temperature to improve the algorithmic performance at a lower temperature. The method is suitable for canonical as well as for generalized ensemble simulations. A simple approximation to the full transformation is tested at room temperature for Met-Enkephalin in vacuum. Integrated autocorrelation times are found to be reduced by factors close to two and a similar improvement due to generalized ensemble methods enters multiplicatively. △ Less

Submitted 7 August, 2003; v1 submitted 24 June, 2003; originally announced June 2003.

Comments: Plenary talk at the Los Alamos conference, The Monte Carlo Method in Physical Sciences: Celebrating the 50th Anniversary of the Metropolis Algorithm, to appear in the proceedings, 11 pages, 4 figures, one table. Inconsistencies corrected and references added

Journal ref: AIP Conference Proceedings Vol.690, 63-73 (2003)

Metropolis simulations of Met-Enkephalin with solvent-accessible area parameterizations

Authors: Bernd A. Berg, Hsiao-Ping Hsu

Abstract: We investigate the solvent-accessible area method by means of Metropolis simulations of the brain peptide Met-Enkephalin at 300$ K$. For the energy function ECEPP/2 nine atomic solvation parameter (ASP) sets are studied. The simulations are compared with one another, with simulations with a distance dependent electrostatic permittivity $ε(r)$, and with vacuum simulations ($ε=2$). Parallel temper… ▽ More We investigate the solvent-accessible area method by means of Metropolis simulations of the brain peptide Met-Enkephalin at 300$ K$. For the energy function ECEPP/2 nine atomic solvation parameter (ASP) sets are studied. The simulations are compared with one another, with simulations with a distance dependent electrostatic permittivity $ε(r)$, and with vacuum simulations ($ε=2$). Parallel tempering and the biased Metropolis techniques RM$_1$ are employed and their performance is evaluated. The measured observables include energy and dihedral probability densities (pds), integrated autocorrelation times, and acceptance rates. Two of the ASP sets turn out to be unsuitable for these simulations. For all other systems selected configurations are minimized in search of the global energy minima, which are found for the vacuum and the $ε(r)$ system, but for none of the ASP models. Other observables show a remarkable dependence on the ASPs. In particular, we find three ASP sets for which the autocorrelations at 300$ $K are considerably smaller than for vacuum simulations. △ Less

Submitted 17 June, 2003; originally announced June 2003.

Comments: 10 pages and 8 figures

Journal ref: Phys. Rev. E 69, 026703 (2004)

Multi-Overlap Simulations for Transitions between Reference Configurations

Authors: B. A. Berg, H. Noguchi, Y. Okamoto

Abstract: We introduce a new procedure to construct weight factors, which flatten the probability density of the overlap with respect to some pre-defined reference configuration. This allows one to overcome free energy barriers in the overlap variable. Subsequently, we generalize the approach to deal with the overlaps with respect to two reference configurations so that transitions between them are induce… ▽ More We introduce a new procedure to construct weight factors, which flatten the probability density of the overlap with respect to some pre-defined reference configuration. This allows one to overcome free energy barriers in the overlap variable. Subsequently, we generalize the approach to deal with the overlaps with respect to two reference configurations so that transitions between them are induced. We illustrate our approach by simulations of the brainpeptide Met-enkephalin with the ECEPP/2 energy function using the global-energy-minimum and the second lowest-energy states as reference configurations. The free energy is obtained as functions of the dihedral and the root-mean-square distances from these two configurations. The latter allows one to identify the transition state and to estimate its associated free energy barrier. △ Less

Submitted 3 May, 2003; originally announced May 2003.

Comments: 12 pages, (RevTeX), 14 figures, Phys. Rev. E, submitted

Report number: IMS-TH-03-2

Metropolis Importance Sampling for Rugged Dynamical Variables

Authors: Bernd A. Berg

Abstract: A funnel transformation is introduced, which acts recursively from higher towards lower temperatures. It biases the a-priori probabilities of a canonical or generalized ensemble Metropolis simulation, so that they zoom in on the global energy minimum, if a funnel exists indeed. A first, crude approximation to the full transformation, called rugged Metropolis one (RM$_1$), is tested for Met-Enkep… ▽ More A funnel transformation is introduced, which acts recursively from higher towards lower temperatures. It biases the a-priori probabilities of a canonical or generalized ensemble Metropolis simulation, so that they zoom in on the global energy minimum, if a funnel exists indeed. A first, crude approximation to the full transformation, called rugged Metropolis one (RM$_1$), is tested for Met-Enkephalin. At 300$ $K the computational gain is a factor of two and, due to its simplicity, RM$_1$ is well suited to replace the conventional Metropolis updating for these kind of systems. △ Less

Submitted 23 March, 2003; v1 submitted 17 September, 2002; originally announced September 2002.

Comments: 4 pages RevTex, 2 figures, 1 table. Revised after referee repots, 400K acceptance rates corrected

Journal ref: Phys. Rev. Lett. 90, 180601 (2003)

Multicanonical Simulations Step by Step

Authors: Bernd A. Berg

Abstract: The purpose of this article is to provide a starter kit for multicanonical simulations in statistical physics. Fortran code for the $q$-state Potts model in $d=2, 3,...$ dimensions can be downloaded from the Web and this paper describes simulation results, which are in all details reproducible by running prepared programs. To allow for comparison with exact results, the internal energy, the spec… ▽ More The purpose of this article is to provide a starter kit for multicanonical simulations in statistical physics. Fortran code for the $q$-state Potts model in $d=2, 3,...$ dimensions can be downloaded from the Web and this paper describes simulation results, which are in all details reproducible by running prepared programs. To allow for comparison with exact results, the internal energy, the specific heat, the free energy and the entropy are calculated for the $d=2$ Ising ($q=2$) and the $q=10$ Potts model. % in a temperature range from $T=\infty$ down to sufficiently low % temperatures, such that the groundstates are included in the sampling. Analysis programs, relying on an all-log jackknife technique, which is suitable for handling sums of very large numbers, are introduced to calculate our final estimators. △ Less

Submitted 9 August, 2002; v1 submitted 18 June, 2002; originally announced June 2002.

Journal ref: Comput.Phys.Commun. 153 (2003) 397-406

Overlap Distribution of the Three-Dimensional Ising Model

Authors: Bernd A. Berg, Alain Billoire, Wolfhard Janke

Abstract: We study the Parisi overlap probability density P_L(q) for the three-dimensional Ising ferromagnet by means of Monte Carlo (MC) simulations. At the critical point P_L(q) is peaked around q=0 in contrast with the double peaked magnetic probability density. We give particular attention to the tails of the overlap distribution at the critical point, which we control over up to 500 orders of magnitu… ▽ More We study the Parisi overlap probability density P_L(q) for the three-dimensional Ising ferromagnet by means of Monte Carlo (MC) simulations. At the critical point P_L(q) is peaked around q=0 in contrast with the double peaked magnetic probability density. We give particular attention to the tails of the overlap distribution at the critical point, which we control over up to 500 orders of magnitude by using the multi-overlap MC algorithm. Below the critical temperature interface tension estimates from the overlap probability density are given and their approach to the infinite volume limit appears to be smoother than for estimates from the magnetization. △ Less

Submitted 17 May, 2002; originally announced May 2002.

Comments: 7 pages, RevTex, 9 Postscript figures

Report number: Leipzig-LU-ITP 2002/008

Journal ref: Phys. Rev. E 66, 046122 (2002)

Multi-overlap simulations of spin glasses

Authors: Wolfhard Janke, Bernd A. Berg, Alain Billoire

Abstract: We present results of recent high-statistics Monte Carlo simulations of the Edwards-Anderson Ising spin-glass model in three and four dimensions. The study is based on a non-Boltzmann sampling technique, the multi-overlap algorithm which is specifically tailored for sampling rare-event states. We thus concentrate on those properties which are difficult to obtain with standard canonical Boltzmann… ▽ More We present results of recent high-statistics Monte Carlo simulations of the Edwards-Anderson Ising spin-glass model in three and four dimensions. The study is based on a non-Boltzmann sampling technique, the multi-overlap algorithm which is specifically tailored for sampling rare-event states. We thus concentrate on those properties which are difficult to obtain with standard canonical Boltzmann sampling such as the free-energy barriers F^q_B in the probability density P_J(q) of the Parisi overlap parameter q and the behaviour of the tails of the disorder averaged density P(q) = [P_J(q)]_av. △ Less

Submitted 3 December, 2001; originally announced December 2001.

Comments: 14 pages, Latex, 18 Postscript figures, to be published in NIC Series - Publication Series of the John von Neumann Institute for Computing (NIC)

Report number: Leipzig-LU-ITP 2001/026, Saclay-T01/132

Journal ref: NIC Series - Publication Series of the John von Neumann Institute for Computing, Jülich, 9, 301-314 (2002)

Generalized Ensemble Simulations for Complex Systems

Authors: Bernd A. Berg

Abstract: The most efficient MC weights for the calculation of physical, canonical expectation values are not necessarily those of the canonical ensemble. The use of suitably generalized ensembles can lead to a much faster convergence of the simulation. Although not realized by nature, these ensembles can be implemented on computers. In recent years generalized ensembles have in particular been studied fo… ▽ More The most efficient MC weights for the calculation of physical, canonical expectation values are not necessarily those of the canonical ensemble. The use of suitably generalized ensembles can lead to a much faster convergence of the simulation. Although not realized by nature, these ensembles can be implemented on computers. In recent years generalized ensembles have in particular been studied for the simulation of complex systems. For these systems it is typical that conflicting constraints lead to free energy barriers, which fragment the configuration space. Examples of major interest are spin glasses and proteins. In my overview I first comment on the strengths and weaknesses of a few major approaches, multicanonical simulations, transition variable methods, and parallel tempering. Subsequently, two applications are presented: a new analysis of the Parisi overlap distribution for the 3d Edwards-Anderson Ising spin glass and the helix-coil transition of amino-acid homo-oligomers. △ Less

Submitted 5 June, 2002; v1 submitted 24 October, 2001; originally announced October 2001.

Comments: Plenary talk at the CCP 2001 conference, submitted to Comp. Phys. Commun., 7 pages, 3 figures, some typos eliminated

Journal ref: Comp. Phys. Commun 147 (2002) 52

About the Functional Form of the Parisi Overlap Distribution for the Three-Dimensional Edwards-Anderson Ising Spin Glass

Authors: Bernd A. Berg, Alain Billoire, Wolfhard Janke

Abstract: Recently, it has been conjectured that the statistics of extremes is of relevance for a large class of correlated system. For certain probability densities this predicts the characteristic large $x$ fall-off behavior $f(x)\sim\exp (-a e^x)$, $a>0$. Using a multicanonical Monte Carlo technique, we have calculated the Parisi overlap distribution $P(q)$ for the three-dimensional Edward-Anderson Isi… ▽ More Recently, it has been conjectured that the statistics of extremes is of relevance for a large class of correlated system. For certain probability densities this predicts the characteristic large $x$ fall-off behavior $f(x)\sim\exp (-a e^x)$, $a>0$. Using a multicanonical Monte Carlo technique, we have calculated the Parisi overlap distribution $P(q)$ for the three-dimensional Edward-Anderson Ising spin glass at and below the critical temperature, even where $P(q)$ is exponentially small. We find that a probability distribution related to extreme order statistics gives an excellent description of $P(q)$ over about 80 orders of magnitude. △ Less

Submitted 2 August, 2001; originally announced August 2001.

Comments: 4 pages RevTex, 3 figures

Journal ref: Phys. Rev. E65, 045102R 2002

Spin glass overlap barriers in three and four dimensions

Authors: Bernd A. Berg, Alain Billoire, Wolfhard Janke

Abstract: For the Edwards-Anderson Ising spin-glass model in three and four dimensions (3d and 4d) we have performed high statistics Monte Carlo calculations of those free-energy barriers $F^q_B$ which are visible in the probability density $P_J(q)$ of the Parisi overlap parameter $q$. The calculations rely on the recently introduced multi-overlap algorithm. In both dimensions, within the limits of lattic… ▽ More For the Edwards-Anderson Ising spin-glass model in three and four dimensions (3d and 4d) we have performed high statistics Monte Carlo calculations of those free-energy barriers $F^q_B$ which are visible in the probability density $P_J(q)$ of the Parisi overlap parameter $q$. The calculations rely on the recently introduced multi-overlap algorithm. In both dimensions, within the limits of lattice sizes investigated, these barriers are found to be non-self-averaging and the same is true for the autocorrelation times of our algorithm. Further, we present evidence that barriers hidden in $q$ dominate the canonical autocorrelation times. △ Less

Submitted 15 April, 2000; v1 submitted 21 October, 1999; originally announced October 1999.

Comments: 20 pages, Latex, 12 Postscript figures, revised version to appear in Phys. Rev. B

Report number: Saclay-T99/095

Journal ref: Phys. Rev. B61 (2000) 12143-12150

Introduction to Multicanonical Monte Carlo Simulations

Authors: Bernd A. Berg

Abstract: Monte Carlo simulation with {\it a-priori} unknown weights have attracted recent attention and progress has been made in understanding (i) the technical feasibility of such simulations and (ii) classes of systems for which such simulations lead to major improvements over conventional Monte Carlo simulations. After briefly sketching the history of multicanonical calculations and their range of ap… ▽ More Monte Carlo simulation with {\it a-priori} unknown weights have attracted recent attention and progress has been made in understanding (i) the technical feasibility of such simulations and (ii) classes of systems for which such simulations lead to major improvements over conventional Monte Carlo simulations. After briefly sketching the history of multicanonical calculations and their range of application, a general introduction in the context of the statistical physics of the d-dimensional generalized Potts models is given. Multicanonical simulations yield canonical expectation values for a range of temperatures or any other parameter(s) for which appropriate weights can be constructed. We shall address in some details the question how the multicanonical weights are actually obtained. Subsequently miscellaneous topics related to the considered algorithms are reviewed. Then multicanonical studies of first order phase transitions are discussed and finally applications to complex systems such as spin glasses and proteins. △ Less

Submitted 15 September, 1999; originally announced September 1999.

Comments: Review, 24 pages

Journal ref: Fields Inst.Commun.26:1,2000

Quantum Chaos in Compact Lattice QED

Authors: B. A. Berg, H. Markum, R. Pullirsch

Abstract: Complete eigenvalue spectra of the staggered Dirac operator in quenched $4d$ compact QED are studied on $8^3 \times 4$ and $8^3 \times 6$ lattices. We investigate the behavior of the nearest-neighbor spacing distribution $P(s)$ as a measure of the fluctuation properties of the eigenvalues in the strong coupling and the Coulomb phase. In both phases we find agreement with the Wigner surmise of th… ▽ More Complete eigenvalue spectra of the staggered Dirac operator in quenched $4d$ compact QED are studied on $8^3 \times 4$ and $8^3 \times 6$ lattices. We investigate the behavior of the nearest-neighbor spacing distribution $P(s)$ as a measure of the fluctuation properties of the eigenvalues in the strong coupling and the Coulomb phase. In both phases we find agreement with the Wigner surmise of the unitary ensemble of random-matrix theory indicating quantum chaos. Combining this with previous results on QCD, we conjecture that quite generally the non-linear couplings of quantum field theories lead to a chaotic behavior of the eigenvalues of the Dirac operator. △ Less

Submitted 10 December, 1998; originally announced December 1998.

Comments: 11 pages, 4 figures

Journal ref: Phys.Rev. D59 (1999) 097504

Energy barriers of spin glasses from multi-overlap simulations

Authors: Wolfhard Janke, Bernd A. Berg, Alain Billoire

Abstract: We report large-scale simulations of the three-dimensional Edwards-Anderson Ising spin glass system using the recently introduced multi-overlap Monte Carlo algorithm. In this approach the temperature is fixed and two replica are coupled through a weight factor such that a broad distribution of the Parisi overlap parameter $q$ is achieved. Canonical expectation values for the entire $q$-range (mu… ▽ More We report large-scale simulations of the three-dimensional Edwards-Anderson Ising spin glass system using the recently introduced multi-overlap Monte Carlo algorithm. In this approach the temperature is fixed and two replica are coupled through a weight factor such that a broad distribution of the Parisi overlap parameter $q$ is achieved. Canonical expectation values for the entire $q$-range (multi-overlap) follow by reweighting. We present an analysis of the performance of the algorithm and in particular discuss results on spin glass free-energy barriers which are hard to obtain with conventional algorithms. In addition we discuss the non-trivial scaling behavior of the canonical $q$-distributions in the broken phase. △ Less

Submitted 30 November, 1998; originally announced November 1998.

Comments: 10 pages, Latex, 9 Postscript figures

Report number: NTZ 29/1998

Journal ref: Ann. Phys. (Leipzig) 7 (1998) 544

Configuration Space for Random Walk Dynamics

Authors: Bernd A. Berg, Ulrich H. E. Hansmann

Abstract: Applied to statistical physics models, the random cost algorithm enforces a Random Walk (RW) in energy (or possibly other thermodynamic quantities). The dynamics of this procedure is distinct from fixed weight updates. The probability for a configuration to be sampled depends on a number of unusual quantities, which are explained in this paper. This has been overlooked in recent literature, wher… ▽ More Applied to statistical physics models, the random cost algorithm enforces a Random Walk (RW) in energy (or possibly other thermodynamic quantities). The dynamics of this procedure is distinct from fixed weight updates. The probability for a configuration to be sampled depends on a number of unusual quantities, which are explained in this paper. This has been overlooked in recent literature, where the method is advertised for the calculation of canonical expectation values. We illustrate these points for the $2d$ Ising model. In addition, we proof a previously conjectured equation which relates microcanonical expectation values to the spectral density. △ Less

Submitted 10 June, 1998; v1 submitted 13 May, 1998; originally announced May 1998.

Comments: Various minor changes, appendix added, Fig. 2 dropped

Report number: MTU-PHY-H1

Multi-Overlap Simulations of the $3d$ Edwards-Anderson Ising Spin Glass

Authors: Bernd A. Berg, Wolfhard Janke

Abstract: We introduce a novel method for numerical spin glass investigations: Simulations of two replica at fixed temperature, weighted such that a broad distribution of the Parisi overlap parameter $q$ is achieved. Canonical expectation values for the entire $q$-range (multi-overlap) follow by re-weighting. We demonstrate the feasibility of the approach by studying the $3d$ Edwards-Anderson Ising (… ▽ More We introduce a novel method for numerical spin glass investigations: Simulations of two replica at fixed temperature, weighted such that a broad distribution of the Parisi overlap parameter $q$ is achieved. Canonical expectation values for the entire $q$-range (multi-overlap) follow by re-weighting. We demonstrate the feasibility of the approach by studying the $3d$ Edwards-Anderson Ising ($J_{ik}=\pm 1$) spin glass in the broken phase ($β=1$). For the first time it becomes possible to obtain reliable results about spin glass tunneling barriers. In addition, as do some earlier numerical studies, our results support that Parisi mean field theory is valid down to $3d$. △ Less

Submitted 30 December, 1997; originally announced December 1997.

Comments: 10 pages latex and 3 postscript figures in one file

Journal ref: Phys.Rev.Lett.80:4771-4774,1998

Algorithmic aspects of multicanonical simulations

Authors: Bernd A. Berg

Abstract: Monte Carlo (MC) simulations of many systems, in particular those with conflicting constraints, can be considerably speeded up by using multicanonical or related methods. Some of these approaches sample with a-priori unknown weight factors. After introducing the concept, I shall focus on two aspects: (i) Opinions about the optimal choice of weight factors. (ii) Methods to get weight factor estim… ▽ More Monte Carlo (MC) simulations of many systems, in particular those with conflicting constraints, can be considerably speeded up by using multicanonical or related methods. Some of these approaches sample with a-priori unknown weight factors. After introducing the concept, I shall focus on two aspects: (i) Opinions about the optimal choice of weight factors. (ii) Methods to get weight factor estimates, with emphasize on a multicanonical recursion. △ Less

Submitted 5 August, 1997; originally announced August 1997.

Comments: 4 pages latex, Slightly expanded contribution to the LAT97 Edinburgh conference, to appear in the proceedings, Nucl. Phys. B (Proc. Suppl.)

Journal ref: Nucl.Phys.Proc.Suppl. 63 (1998) 982-984

Brief Review of Multicanonical Simulations

Authors: Bernd A. Berg

Abstract: Recent progress of simulations with non-canonical weight factors is summarized. Recent progress of simulations with non-canonical weight factors is summarized. △ Less

Submitted 7 May, 1997; originally announced May 1997.

Comments: 11 pages latex and sprocl.sty. Published in the Proceedings of the International Conference on Multiscale Phenomena and Their Simulations (Bielefeld, October 1996), edited by F. Karsch, B. Monien and H. Satz (World Scientific, 1997)

Multicanonical Recursions

Authors: Bernd A. Berg

Abstract: The problem of calculating multicanonical parameters recursively is discussed. I describe in detail a computational implementation which has worked reasonably well in practice. The problem of calculating multicanonical parameters recursively is discussed. I describe in detail a computational implementation which has worked reasonably well in practice. △ Less

Submitted 17 March, 1995; v1 submitted 16 March, 1995; originally announced March 1995.

Comments: 23 pages, latex, 4 postscript figures included (uuencoded Z-compressed .tar file created by uufiles), figure file corrected.

Report number: SCRI 95-29

Journal ref: J.Statist.Phys. 82 (1996) 323-342

Grundstate Properties of the 3D Ising Spin Glass

Authors: Bernd A. Berg, Ulrich H. E. Hansmann, Tarik Celik

Abstract: We study zero--temperature properties of the 3d Edwards--Anderson Ising spin glass on finite lattices up to size $12^3$. Using multicanonical sampling we generate large numbers of groundstate configurations in thermal equilibrium. Finite size scaling with a zero--temperature scaling exponent $y = 0.74 \pm 0.12$ describes the data well. Alternatively, a descriptions in terms of Parisi mean field… ▽ More We study zero--temperature properties of the 3d Edwards--Anderson Ising spin glass on finite lattices up to size $12^3$. Using multicanonical sampling we generate large numbers of groundstate configurations in thermal equilibrium. Finite size scaling with a zero--temperature scaling exponent $y = 0.74 \pm 0.12$ describes the data well. Alternatively, a descriptions in terms of Parisi mean field behaviour is still possible. The two scenarios give significantly different predictions on lattices of size $\ge 12^3$. △ Less

Submitted 6 December, 1993; originally announced December 1993.

Comments: LATEX 9pages,figures upon request ,SCRI-93

A Determination of Interface Free Energies

Authors: A. Billoire, T. Neuhaus, B. Berg

Abstract: We determine the interface free energy $F_{o.d.}$ between disordered and ordered phases in the q=10 and q=20 2-d Potts models using the results of multicanonical Monte Carlo simulations on $L^2$ lattices, and suitable finite volume estimators. Our results, when extrapolated to the infinite volume limit, agree to high precision with recent analytical calculations. At the transition point $β_t$ th… ▽ More We determine the interface free energy $F_{o.d.}$ between disordered and ordered phases in the q=10 and q=20 2-d Potts models using the results of multicanonical Monte Carlo simulations on $L^2$ lattices, and suitable finite volume estimators. Our results, when extrapolated to the infinite volume limit, agree to high precision with recent analytical calculations. At the transition point $β_t$ the probability distribution function of the energy exhibits two maxima. Their locations have $1/L^2$ corrections, in contradiction with claims of $1/L$ behavior made in the literature. Our data show a flat region inbetween the two maxima which characterizes two domain configurations. △ Less

Submitted 30 August, 1993; v1 submitted 26 July, 1993; originally announced July 1993.

Comments: Submited to Nuclear Physics B (FS) Latex file, 24 pages, 11 PostScript figures. Saclay preprint SPhT-93/65

Journal ref: Nucl.Phys. B413 (1994) 795-812

Study of +/-J Ising Spin Glasses via Multicanonical Ensemble

Authors: Tarik Celik, Ulrich H. E. Hansmann, Bernd Berg

Abstract: We performed numerical simulations of 2D and 3D Edwards-Anderson spin glass models by using the recently developed multicanonical ensemble. Our ergodicity times increase with the lattice size approximately as $V^3$. The energy, entropy and other physical quantities are easily calculable at all temperatures from a single simulation. Their finite size scalings and the zero temperature limits are a… ▽ More We performed numerical simulations of 2D and 3D Edwards-Anderson spin glass models by using the recently developed multicanonical ensemble. Our ergodicity times increase with the lattice size approximately as $V^3$. The energy, entropy and other physical quantities are easily calculable at all temperatures from a single simulation. Their finite size scalings and the zero temperature limits are also explored. △ Less

Submitted 12 March, 1993; originally announced March 1993.

Comments: no figures included, 5 pages, FSU-SCRI-93-31

Observation of FSS for a First Order Phase Transition

Authors: A. Billoire, T. Neuhaus, B. Berg

Abstract: We present the results of a multicanonical simulation of the q=20 2-d Potts model in the transition region. This is a very strong first order phase transition. We observe, for the first time, the asymptotic finite size scaling behavior predicted by Borgs and Kotecký close to a first order phase transition point. We present the results of a multicanonical simulation of the q=20 2-d Potts model in the transition region. This is a very strong first order phase transition. We observe, for the first time, the asymptotic finite size scaling behavior predicted by Borgs and Kotecký close to a first order phase transition point. △ Less

Submitted 5 November, 1992; originally announced November 1992.

Comments: Submited to Nuclear Physics B (FS) Latex file, 15 pages, 5 PostScript figures. Saclay preprint SPhT-92/120

Journal ref: Nucl.Phys. B396 (1993) 779-788

Multicanonical Study of the 3D Ising Spin Glass

Authors: B. A. Berg, T. Celik, U. Hansmann

Abstract: We simulated the Edwards-Anderson Ising spin glass model in three dimensions via the recently proposed multicanonical ensemble. Physical quantities such as energy density, specific heat and entropy are evaluated at all temperatures. We studied their finite size scaling, as well as the zero temperature limit to explore the ground state properties. We simulated the Edwards-Anderson Ising spin glass model in three dimensions via the recently proposed multicanonical ensemble. Physical quantities such as energy density, specific heat and entropy are evaluated at all temperatures. We studied their finite size scaling, as well as the zero temperature limit to explore the ground state properties. △ Less

Submitted 7 October, 1992; originally announced October 1992.

Comments: FSU-SCRI-92-121; 7 pages; sorry, no figures included

Journal ref: Nucl.Phys.Proc.Suppl. 34 (1993) 664-665

Multicanonical Spin Glass Simulations

Authors: B. A. Berg, T. Celik

Abstract: We report a Monte Carlo simulation of the $2D$ Edwards-Anderson spin glass model within the recently introduced multicanonical ensemble. Replica on lattices of size $L^2$ up to $L=48$ are investigated. Once a true groundstate is found, we are able to give a lower bound on the number of statistically independent groundstates sampled. Temperature dependence of the energy, entropy and other quantit… ▽ More We report a Monte Carlo simulation of the $2D$ Edwards-Anderson spin glass model within the recently introduced multicanonical ensemble. Replica on lattices of size $L^2$ up to $L=48$ are investigated. Once a true groundstate is found, we are able to give a lower bound on the number of statistically independent groundstates sampled. Temperature dependence of the energy, entropy and other quantities of interest are easily calculable. In particular we report the groundstate results. Computations involving the spin glass order parameter are more tedious. Our data indicate that the large $L$ increase of the ergodicity time is reduced to an approximately $V^3$ power law. Altogether the results suggest that the multicanonical ensemble improves the situation of simulations for spin glasses and other systems which have to cope with similar problems of conflicting constraints. △ Less

Submitted 21 July, 1992; originally announced July 1992.

Comments: 24 pages

Report number: FSU-SCRI-92-75

Journal ref: Int.J.Mod.Phys. C3 (1992) 1251-1274