基于谓词逻辑和约束网络结构的遗传编程方法

GENETIC PROGRAMMING BASED ON PREDICATE LOGIC AND CONSTRAINT NETWORK ARCHITECTURE

提出了一种新的谓词型遗传编程方法,该方法包括:待求解问题论域的函数空间约束、遗传算子的谓词逻辑表达、谓词结构推理下数值参数空间的混沌式遍历及发生机制以及非线性方式下自适应进化和非监督学习的综合。整个过程是在约束网络的基础上,以谓词逻辑表达机制为核心实现的函数结点集合约束下的数值型进化计算过程。该方法可被认为是对现有遗传编程技术的改进,克服了LISP型及其C仿真型遗传编程算法在各种表示手段方面的缺陷,提高了计算效率,在系统辨识中的应用实验表明了是有效和实用的。

This paper proposes a novel predicate genetic programming method. It mainly includes: (1) functional space constraint aimed at a specific problem to be solved; (2) predicate logic representation of genetic operator; (3) chaotic ergod and generating mechanism of valued type of parameter space with predicate architecture inference; (4) synthesis of self-adaptive evolution and unsupervised learning under nonlinear mode. The whole process is a valued type of evolutionary computational process realized under the constraint of function node set based on constraint network architecture. The kernel of the computation is constructed by predicate logic representation mechanism. This method can be regarded as an improvement on the existing genetic programming techniques. The disadvantage of LISP type and its C simulating type of genetic programming algorithms have been avoided and computational efficiency raised. These can be expressed in the following three aspects: (1) the predicate GP is more efficient and flexible than the existing GP methods; (2) the specific domain process constraint can be efficiently represented, which is reflected in the nonlinear mapping of genetic operator window selection under predicate representation architecture; (3) with the embeded constraint network, it makes the evolutionary computational mechanism possess the functions of self-adaptive architecture-modification and self-learning. Computer simulating result shows the advantage and practical value of the method.