A Theory on Flat Histogram Monte Carlo Algorithms
نویسنده
چکیده
The flat histogram Monte Carlo algorithms have been successfully used in many problems in scientific computing.However, there is no a rigorous theory for the convergence of the algorithms. In this paper, a modified flat histogram algorithm is presented and its convergence is studied. The convergence of the multicanonical algorithm and the Wang-Landau algorithm is argued based on their relations to the modified algorithm. The numerical results show the superiority of the modified algorithm to the multicanonical and Wang-Landau algorithms.
منابع مشابه
Transition matrix Monte Carlo and flat-histogram algorithm
In any valid Monte Carlo sampling that realizes microcanonical property we can collect statistics for a transition matrix in energy. This matrix is used to determine the density of states, from which most of the thermodynamical averages can be calculated, including free energy and entropy. We discuss single-spin-flip algorithms, such as flat-histogram and equal-hit algorithms, that can be used ...
متن کاملFlat histogram Monte Carlo method
We discuss a sampling algorithm which generates flat histogram in energy. In combination with transition matrix Monte Carlo, the density of states and derived quantities such as entropy and free energy as a function of temperature can be computed in a single simulation.
متن کاملEfficient Monte Carlo Simulation Methods in Statistical Physics
The basic problem in equilibrium statistical mechanics is to compute phase space average, in which Monte Carlo method plays a very important role. We begin with a review of nonlocal algorithms for Markov chain Monte Carlo simulation in statistical physics. We discuss their advantages, applications, and some challenge problems which are still awaiting for better solutions. We discuss some of the...
متن کاملA flat histogram method for inference with probabilistic and deterministic constraints
Reasoning in a context where both probabilistic and deterministic dependencies are present at the same time is a challenging task with many real-world applications. Classic inference methods like Gibbs Sampling and Message Passing algorithms tend to give poor results in the presence of deterministic constraints, while purely logical reasoning techniques are not designed for probabilistic depend...
متن کاملOn the Pseudo Phase Diagram of Single Semi-Flexible Polymer Chains: A Flat-Histogram Monte Carlo Study
Abstract: Local stiffness of polymer chains is instrumental in all structure formation processes of polymers, from crystallization of synthetic polymers to protein folding and DNA compactification. We present Stochastic Approximation Monte Carlo simulations—a type of flat-histogram Monte Carlo method—determining the density of states of a model class of single semi-flexible polymer chains, and,...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2004