一种协方差矩阵的多尺度量子谐振子算法

发布时间：2018-05-29 18:28

本文选题：全局优化 + 量子谐振子　；参考：《控制与决策》2017年12期

【摘要】：针对多尺度量子谐振子算法在处理高维全局优化问题时难以收敛的问题,提出一种协方差矩阵的多尺度量子谐振子优化算法,并给出新算法的核心数学模型.所提算法改进了多元正态分布评估算法中的协方差矩阵生成方式,保留了之前采样点的记忆,加入动态迭代步长加快了新协方差矩阵的更新速度.实验结果表明,所提算法的性能远超原算法,与4种经典优化算法相比,在收敛精度、收敛速度和鲁棒性上也具有优势.
[Abstract]:In order to solve the problem that the multi-scale quantum harmonic oscillator algorithm is difficult to converge when dealing with the high-dimensional global optimization problem, a multi-scale quantum harmonic oscillator optimization algorithm based on covariance matrix is proposed, and the core mathematical model of the new algorithm is given. The proposed algorithm improves the method of covariance matrix generation in the multivariate normal distribution evaluation algorithm, retains the memory of the previous sampling points, and adds the dynamic iteration step to accelerate the updating speed of the new covariance matrix. The experimental results show that the performance of the proposed algorithm is far more than the original algorithm. Compared with the four classical optimization algorithms, the proposed algorithm also has advantages in convergence accuracy, convergence speed and robustness.
【作者单位】：成都信息工程大学软件工程学院;西南民族大学计算机科学与技术学院;
【基金】：国家自然科学基金项目(71673032,60702075) 国家社会科学基金项目(12XSH019)
【分类号】：TP18

【相似文献】