量子进化策略

本文将进化策略和量子理论相结合,提出一种新的学习算法-量子进化策略(Quantum Evolutionary Strategies)算法.它是一种基于量子计算的概念和理论(诸如量子比特和量子叠加态)的进化策略算法,在这一算法中,采用量子编码来表征染色体,使用量子变异实现染色体的进化.由于量子变异中融入了当前最优解的信息,同时采用“全干扰交叉”操作克服早熟现象的发生,因此它比传统进化策略具有更快的收敛速度和全局寻优的能力.本文不仅从理论上证明了它的全局收敛性,而且仿真计算也表明了此算法的优越性.

The Quantum Evolutionary Strategies

In this paper,a novel kind of algorithm,the quantum evolutionary strategies-QES,is proposed based on the combination of quantum theory and evolutionary theory.It is a kind of evolutionary strategies with the form of quantum chromosome,whose core lies on the concept and principles of quantum computing,such as qubits and superposition of states.By using qubit mutation,we can make full use of the information of the currently best individual to perform the next search,and use whole interference to avoid prematurity,so it has rapid convergence and good global search capacity.The paper not only proves the global convergence of the QES,but some simulated experimentats are given to prove its superiority to other algorithms.