The Estimation of Distributions and the Minimum Relative Entropy Principle
نویسندگان
چکیده
Estimation of Distribution Algorithms (EDA) have been proposed as an extension of genetic algorithms. In this paper we explain the relationship of EDA to algorithms developed in statistics, artificial intelligence, and statistical physics. The major design issues are discussed within a general interdisciplinary framework. It is shown that maximum entropy approximations play a crucial role. All proposed algorithms try to minimize the Kullback-Leibler divergence KLD between the unknown distribution p(x) and a class q(x) of approximations. However, the Kullback-Leibler divergence is not symmetric. Approximations which suppose that the function to be optimized is additively decomposed (ADF) minimize KLD(q||p), the methods which learn the approximate model from data minimize KLD(p||q). This minimization is identical to maximizing the log-likelihood. In the paper three classes of algorithms are discussed. FDA uses the ADF to compute an approximate factorization of the unknown distribution. The factors are marginal distributions, whose values are computed from samples. The second class is represented by the Bethe-Kikuchi approach which has recently been rediscovered in statistical physics. Here the values of the marginals are computed from a difficult constrained minimization problem. The third class learns the factorization from the data. We analyze our learning algorithm LFDA in detail. It is shown that learning is faced with two problems: first, to detect the important dependencies between the variables, and second, to create an acyclic Bayesian network of bounded clique size.
منابع مشابه
A Note on the Bivariate Maximum Entropy Modeling
Let X=(X1 ,X2 ) be a continuous random vector. Under the assumption that the marginal distributions of X1 and X2 are given, we develop models for vector X when there is partial information about the dependence structure between X1 and X2. The models which are obtained based on well-known Principle of Maximum Entropy are called the maximum entropy (ME) mo...
متن کاملComparison of entropy generation minimization principle and entransy theory in optimal design of thermal systems
In this study, the relationship among the concepts of entropy generation rate, entransy theory, and generalized thermal resistance to the optimal design of thermal systems is discussed. The equations of entropy and entransy rates are compared and their implications for optimization of conductive heat transfer are analyzed. The theoretical analyses show that based on entropy generation minimizat...
متن کاملModeling of the Maximum Entropy Problem as an Optimal Control Problem and its Application to Pdf Estimation of Electricity Price
In this paper, the continuous optimal control theory is used to model and solve the maximum entropy problem for a continuous random variable. The maximum entropy principle provides a method to obtain least-biased probability density function (Pdf) estimation. In this paper, to find a closed form solution for the maximum entropy problem with any number of moment constraints, the entropy is consi...
متن کاملQuasi-continuous maximum entropy distribution approximation with kernel density
This paper extends maximum entropy estimation of discrete probability distributions to the continuous case. This transition leads to a nonparametric estimation of a probability density function, preserving the maximum entropy principle. Furthermore, the derived density estimate provides a minimum mean integrated square error. In a second step it is shown, how boundary conditions can be included...
متن کاملEstimation of Evolution of Relative Humidity Distribution for Concrete Slabs
Realistic prediction of concrete shrinkage and creep requires the calculation of the distributions of relative humidity at various times. Although the distributions of the relative humidity can be computed by numerical methods from the differential equation for diffusion, simple prediction formulas can facilitate structural analysis. The purpose of this paper is to present a simple formula for ...
متن کاملEstimation of Evolution of Relative Humidity Distribution for Concrete Slabs
Realistic prediction of concrete shrinkage and creep requires the calculation of the distributions of relative humidity at various times. Although the distributions of the relative humidity can be computed by numerical methods from the differential equation for diffusion, simple prediction formulas can facilitate structural analysis. The purpose of this paper is to present a simple formula for ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید
ثبت ناماگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید
ورودعنوان ژورنال:
- Evolutionary computation
دوره 13 1 شماره
صفحات -
تاریخ انتشار 2005