NIQGA改进及基于可变角距离的新型量子进化算法

为了提高量子进化算法的执行效率,在NIQGA算法基础上,通过改进△θi和S(αi,βi)参数表提出了一种改进算法INIQGA.又通过引入量子比特间角距离定义,提出了一种基于可变角距离旋转的量子进化算法QEA-VAR,该算法采用旋转门操作进行种群进化时,依据当前染色体中量子比特|φ〉i与最优解对应基态| 0〉或| 1〉的...

Improvement on NIQGA and novel quantum-inspired evolutionary algorithm based on variable angle-distance

In order to enhance the efficiency of the quantum-inspired evolutionary algorithm,on basis of the original NIQGA algorithm,an improved algorithm(INIQGA) is put forward by revising the parameter table of Δθi and S(αi,βi).Through introducing the definition of angle-distance between qubits,a novel quantum-inspired evolutionary algorithm based on the variable angle-distance rotation strategy(QEA-VAR) is proposed.In the QEA-VAR algorithm,the rotation angle is dynamically chosen according to the angle-distance between the qubit φ〉i in current chromosome and the basis state 0〉 or 1〉 of the optimal solution,when the corresponding rotation gate is performed to evolve the quantum population.The whole process does not need complicated look-up table operation.Compared with previous algorithms based on the look-up table mechanism,the QEA-VAR algorithm is more simple,feasible and comprehensible.Experiments on the well-known 0/1 knapsack problem show that INIQGA has a faster convergence and better profits than NIQGA,and QEA-VAR has even higher performance than INIQGA and other similar evolutionary algorithms like QEA,CGA.This effect is getting more apparent with the increase of the items in 0/1 knapsack problem.