arXiv subjects
Publications and source records attributed to A. C. Maggs.
Quantum annealing is analogous to simulated annealing with a tunneling mechanism substituting for thermal activation. Its performance has been tested in numerical simulation with mixed conclusions. There is a class of optimization problems for which the efficiency can be studied analytically using techniques based on the statistical mechanics of spin glasses.
We study first order quantum phase transitions in mean-field spin glasses. We solve the quantum Random Energy Model using elementary methods and show that at the transition the eigenstate suddenly projects onto the unperturbed ground state and that the gap between the lowest states is exponentially small in the system size. We argue that this is a generic feature of all `Random First Order' models, which includes benchmarks such as random satisfiability. We introduce a two-time instanton to calculate this gap in general, and discuss the consequences for quantum annealing.
We study the simulation of charged systems in the presence of general boundary conditions in a local Monte Carlo algorithm based on a constrained electric field. We firstly show how to implement constant-potential, Dirichlet, boundary conditions by introducing extra Monte Carlo moves to the algorithm. Secondly, we show the interest of the algorithm for studying systems which require anisotropic electrostatic boundary conditions for simulating planar geometries such as membranes.
We study numerically the Casimir interaction between dielectrics in both two and three dimensions. We demonstrate how sparse matrix factorizations enable one to study torsional interactions in three dimensions. In two dimensions we study the full cross-over between non-retarded and retarded interactions as a function of separation. We use constrained factorizations in order to measure the interaction of a particle with a rough dielectric surface and compare with a scaling argument.
In systems composed of water and hydrocarbons Van der Waals-interactions are dominated by the non-retarded, classical (Keesom) part of the Lifshitz-interaction; the interaction is screened by salt and extends over mesoscopic distances of the order of the size of the (micellar) constituents of complex fluids. We show that these interactions are included intrinsically in a recently introduced local Monte Carlo algorithm for simulating electrostatic interactions between charges in the presence of non-homogeneous dielectric media.
We study thermal Casimir and quantum non-retarded Lifshitz interactions between dielectrics in general geometries. We map the calculation of the classical partition function onto a determinant which we discretize and evaluate with the help of Cholesky factorization. The quantum partition function is treated by path integral quantization of a set of interacting dipoles and reduces to a product of determinants. We compare the approximations of pairwise additivity and proximity force with our numerical methods. We propose a ``factorization approximation'' which gives rather good numerical results in the geometries that we study.
We develop a discretized theory of thermal Casimir interactions to numerically calculate the interactions between fluctuating dielectrics. From a constrained partition function we derive a surface free energy, while handling divergences that depend on system size and discretization. We derive analytic results for parallel plate geometry in order to check the convergence of the numerical methods. We use the method to calculate vertical and lateral Casimir forces for a set of grooves.
We introduce a multiscale Monte Carlo algorithm to simulate dense simple fluids. The probability of an update follows a power law distribution in its length scale. The collective motion of clusters of particles requires generalization of the Metropolis update rule to impose detailed balance. We apply the method to the simulation of a Lennard-Jones fluid and show improvements in efficiency over conventional Monte Carlo and molecular dynamics, eliminating hydrodynamic slowing down.
We introduce a constrained energy functional to describe dielectric response. We demonstrate that the local functional is a generalization of the long ranged Marcus energy. Our re-formulation is used to implement a cluster Monte Carlo algorithm for the simulation of dielectric media. The algorithm avoids solving the Poisson equation and remains efficient in the presence of spatial heterogeneity, nonlinearity and scale dependent dielectric properties.
We consider Monte Carlo algorithms for the simulation of charged lattice gases with purely local dynamics. We study the mobility of particles as a function of temperature and show that the poor mobility of particles at low temperatures is due to "trails" or "strings" left behind after particle motion. We introduce modified updates which substantially improve the efficiency of the algorithm in this regime.
We simulate TIP3P water using a constrained Monte Carlo algorithm to generate electrostatic interactions eliminating the need to sum over long ranged Coulomb interactions. We study discretization errors when interpolating charges using splines and Gaussians. We compare our implementation to molecular dynamics and Brownian dynamics codes.
We propose a local, O(N) molecular dynamics algorithm for the simulation of charged systems. The long ranged Coulomb potential is generated by a propagating electric field that obeys modified Maxwell equations. On coupling the electrodynamic equations to an external thermostat we show that the algorithm produces an effective Coulomb potential between particles. On annealing the electrodynamic degrees of freedom the field configuration converges to a solution of the Poisson equation much like the electronic degrees of freedom approach the ground state in ab-initio molecular dynamics.
We consider dynamically constrained Monte-Carlo dynamics and show that this leads to the generation of long ranged effective interactions. This allows us to construct a local algorithm for the simulation of charged systems without ever having to evaluate pair potentials or solve the Poisson equation. We discuss a simple implementation of a charged lattice gas as well as more elaborate off-lattice versions of the algorithm. There are analogies between our formulation of electrostatics and the bosonic Hubbard model in the phase approximation. Cluster methods developed for this model further improve the efficiency of the electrostatics algorithm.
We review a family of local algorithms that permit the simulation of charged particles with purely local dynamics. Molecular dynamics formulations lead to discretizations similar to those of ``particle in cell'' methods in plasma physics. We show how to formulate a local Monte-Carlo algorithm in the presence of the long ranged Coulomb interaction. We compare the efficiencies of our molecular dynamics and Monte-Carlo codes.
This article describes Monte-Carlo algorithms for charged systems using constrained updates for the electric field. The method is generalized to treat inhomogeneous dielectric media, electrolytes via the Poisson-Boltzmann equation and considers the problem of charge and current interpolation for off lattice models. We emphasize the differences between this algorithm and methods based on the electrostatic potential, calculated from the Poisson equation.
We introduce a Monte-Carlo algorithm for the simulation of charged particles moving in the continuum. Electrostatic interactions are not instantaneous as in conventional approaches, but are mediated by a constrained, diffusing electric field on an interpolating lattice. We discuss the theoretical justifications of the algorithm and show that it efficiently equilibrates model polyelectrolytes and polar fluids. In order to reduce lattice artifacts that arise from the interpolation of charges to the grid we implement a local, dynamic subtraction algorithm. This dynamic scheme is completely general and can also be used with other Coulomb codes, such as multigrid based methods.
Charged systems interacting via Coulomb forces can be efficiently simulated by introducing a local, diffusing degree of freedom for the electric field. This paper formulates the continuum electrodynamic equations corresponding to the algorithm and studies the spectrum of fluctuations when these equations are coupled to mobile charges. I compare the calculations with simulations of a charged lattice gas, and study the dynamics of charge and density fluctuations. The algorithm can be understood as a realization of a mechanical model of the ether.
Motivated by recent experiments on DNA torsion-force-extension characteristics we consider the writhing geometry of open stiff molecules. We exhibit a cyclic motion which allows arbitrarily large twisting of the end of a molecule via an activated process. This process is suppressed for forces larger than femto-Newtons which allows us to show that experiments are sensitive to a generalization of the Calugareanu-White formula for the writhe. Using numerical methods we compare this formulation of the writhe with recent analytic calculations.