【摘要】 在经典计算环境下,粗糙集的核属性求解只能逐个验证属性,而量子计算则提供了一种并行计算的可能。经分析,量子计算中的Grover算法适用于粗糙集的核属性求解,但经典Grover算法及Grover算法的一些变体都不能总以100%的概率获得目标分量。该研究首先提出一种将固定相位角度改为由目标分量占比决定的自适应匹配相位角度的改进策略;然后将改进策略应用于粗糙集的核属性求解,提出了一种基于自适应相位匹配量子计算的求核算法。通过理论证明和仿真实验,该算法总能以100%的概率得到粗糙集的核属性,其效率相较于传统求核算法有着平方根的加速。更多还原

【Abstract】 Under the circumstances of classic computing,the core of rough set can only be tested one by one.However,given the situation of quantum computing,the possibility of this being done in parallel is enabled.Theoretical analysis shows shat Grover’s algorithm is suitable for the computation of core.Whereas,the classic Grover’s algorithm and its variations cannot acquire target states with the possibility of 100%.To improve the performance,an adapted strategy,replacing fix-point phase rotation with the self-adaptive phase rotation,is proposed.Furthermore,building on the above-mentioned strategy,an algorithm to compute the core based on self-adaptive phase-matching quantum computing is then fulfilled.Theoretical proofs and stimulating experiments suggest that the new algorithm is able to get the core with 100% possibility,and has square root acceleration.更多还原