Cross-Entropy method: convergence issues for extended implementation
نویسنده
چکیده
The cross-entropy method (CE) developed by R. Rubinstein is an elegant practical principle for simulating rare events. The method approximates the probability of the rare event by means of a family of probabilistic models. The method has been extended to optimization, by considering an optimal event as a rare event. CE works rather good when dealing with deterministic function optimization. Now, it appears that two conditions are needed for a good convergence of the method. First, it is necessary to have a family of models sufficiently flexible for discriminating the optimal events. Indirectly, it appears also that the function to be optimized should be deterministic. The purpose of this paper is to consider the case of partially discriminating model family, and of stochastic functions. It will be shown on simple examples that the CE could fail when relaxing these hypotheses. Alternative improvements of the CE method are investigated and compared on random examples in order to handle this issue.
منابع مشابه
Convergence properties of the cross-entropy method for discrete optimization
We present new theoretical convergence results on the Cross-Entropy method for discrete optimization. Our primary contribution is to show that a popular implementation of the Cross-Entropy method converges, and finds an optimal solution with probability arbitrarily close to 1. We also give necessary conditions and sufficient conditions under which an optimal solution is generated eventually wit...
متن کاملSolving Large MAX CUT Problems Using Cross Entropy & Parametric Minimum Cross Entropy
Two stochastic optimization methods: Cross Entropy (CE) and Parametric Minimum Cross Entropy (PME) were tested against Large Max-Cut problems. The problems were taken from the "The DIMACS Library of Mixed Semidefinite-Quadratic-Linear Programs" challenge web site. The two methods achieved close and even better results than the Best Known Solution. In addition to results presentation, several im...
متن کاملA New Populatoin-based Simulated Annealing Algorithm
In this paper, we propose sequential Monte Carlo simulated annealing (SMC-SA), a populationbased simulated annealing algorithm, for continuous global optimization. SMC-SA incorporates the sequential Monte Carlo method to track the converging sequence of Boltzmann distributions in simulated annealing, such that the empirical distribution will converge weakly to the uniform distribution on the se...
متن کاملMulti-Level Minimum Cross Entropy Thresholding Using Gamma Distribution
Thresholding is one of the popular and fundamental techniques for conducting image segmentation. Many thresholding techniques have been proposed in the literature. Among them, the minimum cross entropy thresholding has been widely adopted. Most minimum cross entropy thresholding methods use Gaussian distribution as an ideal reference histogram for the images to be thresholded. Clearly, it is do...
متن کاملDistributed Asynchronous Algorithm for Cross-Entropy-Based Combinatorial Optimization
Combinatorial optimization algorithms are used in many and diverse applications; for instance, in the planning, management, and operation of manufacturing and logistic systems and communication networks. For scalability and dependability reasons, distributed and asynchronous implementations of these optimization algorithms have obvious advantages over centralized implementations. Several such a...
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:
- CoRR
دوره abs/math/0609461 شماره
صفحات -
تاریخ انتشار 2006