摘要
提出了一种基于实数编码和目标函数梯度信息的量子遗传算法．该方法用量子比特构成染色体,用实数对量子比特进行编码,用量子旋转门进行染色体更新,用量子非门进行染色体变异．对旋转门的旋转角方向的选择,给出了简易快捷的方法；对旋转角大小的选择,结合了目标函数的梯度信息．该方法将每一量子位看作上下两个并列的基因,每条染色体包含两条并列的基因链,每条基因链代表一个优化解．在染色体数目相同时,可使搜索空间加倍．以函数极值问题和神经网络权值优化问题为例,验证了该方法的有效性．
A quantum genetic algorithm based on real number encoding and gradient of object function is presented. In this study, chromosomes are comprised of quantum bits encoded by real number. Chromosomes are renovated by quantum rotating gates and mutated by quantum non - gate. For direction of rotating angle of rotating gate, a convenient method is shown. The gradients of object function are utilized in choosing the value of rotating angle. In this method each quantum bit is regarded as two coordinate genes, each chromosome contains two chain of genes, each chain of genes represents a optimization result. Therefore, a double searching space is acquired for the same number of chromosomes. Finally the availability of the approach is illustrated by two application examples of function extremum and weighting optimization of neural networks.
出处
《哈尔滨工业大学学报》
EI
CAS
CSCD
北大核心
2006年第8期1216-1218,1223,共4页
Journal of Harbin Institute of Technology
基金
国家自然科学基金重点资助项目(50138010)
关键词
遗传算法
量子遗传算法
量子旋转门
实数编码
Genetic algorithm
Quantum genetic algorithm
Quantum rotating gate
Real encoding
