摘要
针对量子进化算法(QEA)求解二进制编码问题比较有效,而求解多进制编码问题则比较困难,提出一种概率进化算法(PEA)。该算法汲取了量子复合位、叠加态等思想,采用由观测概率构成的概率复合位进行编码,观测和更新操作直接针对观测概率进行。PEA保持了QEA的性能,运算速度远优于QEA,并可以采用任意进制编码。函数优化和背包问题实验验证了PEA的有效性。
For Quantum-inspired Evolutionary Algorithm（QEA） is suitable to be used in the problems that use binary coding,but hard be used in that use muhinary coding,a novel evolutionary algorithm called Probability Evolutionary Algorithm（PEA） is presented.PEA dirived from the concepts of quantum bit and superposition of states.The Compound States of probability which is constituted of observing probability is used in PEA.The observing and update method operate the observing probability directly.PEA peforms as better as QEA and runs more fast than QEA.Mulitinary coding can be used in PEA.The function optimization and knapsack problem show the effectiveness of PEA.
出处
《计算机工程与应用》
CSCD
北大核心
2005年第33期64-67,共4页
Computer Engineering and Applications
关键词
量子进化算法概率进化算法
函数优化
背包问题
Quantum-inspired Evolutionary Algorithm,Probability Evolutionary Algorithm, function optimization, knapsack problem
