智能优化方法的统一迭代采样框架
首发时间:2017-05-16
摘要:智能优化方法是解决复杂优化问题的一类通用求解方法,其主要特征是借鉴自然规律建立启发式的解空间搜索/采样策略,通过迭代计算优化问题求解。智能优化方法这一研究方向已经具有60多年的发展历史,目前仍然在蓬勃发展,并在很多研究领域中取得成功应用。但是,无论什么样的智能优化方法,其本质都是一种解空间采样方法。因此,我们可以从采样过程的角度建立智能优化方法的统一迭代采样框架,并在这个框架的基础上,对典型的智能优化方法的迭代过程进行分析和研究。另外,统一的迭代采样框架也有利于系统地分析不同方法的相似性和差异性,从而为有效的算法设计提供指导思想。基于提出的统一迭代采样框架,本文对八种典型智能优化方法的迭代采样过程进行了分析和研究。
For information in English, please click here
A Unifying Iterative Sampling Framework for Intelligent Optimization Methods
Abstract:Intelligent optimization methods (IOMs) are a kind of general problem-solving methods used for complex optimization problems. Inspired by various mechanisms of the nature, IOMs adopt heuristic search/sampling strategies and solve optimization problems by iterative computation. IOMs were born more than 60 years ago, and it is still booming nowadays. Plenty of successful applications have been witnessed in many research fields. However, no matter which IOM, the essence is iterative sampling in solution space. Therefore, we can build a unifying iterative sampling framework for IOMs from the perspective of the sampling process, and analyze and investigate the iteration process of typical IOMs based on the framework. Besides, a unifying iterative sampling framework can help to systematically analyze the similarities and differences between different methods thus providing the guidance for effective algorithm design. Based the proposed iterative sampliing framework, the iterative sampling processes of eight typical intelligent optimization methods are analyzed and studied in this paper.
Keywords: Intelligent optimization unifying framework iteration operator generating operator selection operator
基金:
论文图表:
引用
No.4733518120059614****
同行评议
共计0人参与
勘误表
智能优化方法的统一迭代采样框架
评论
全部评论0/1000