万有引力与群体状态自适应的智能优化算法

Novel Optimization Algorithm Based on Universal Gravitational and Colon State Adaptation

在大数据时代,优化问题的数据量不断变大、数据维度不断变高,并且具有更加复杂的多模性质,因此传统的优化算法对问题的解决表现和分析效果并不能满足人们的要求。启发于群体状态控制过程中会趋于稳定的现象,根据万有引力和四种粒子运动方式,提出一种新的群体优化算法(Colon State Adaptive Optimization Algorithm based on Gravitation,G-CSAO)。在函数极值实验中,提出的算法具有更高的寻优精度,不易陷入局部最优值,能更好地应对多峰问题和高维度问题。将提出的模型与最小距离准则相结合应用在模式分类问题中,UCI数据集实验结果表明,提出的模型具有更好的分类性能及应用前景。

In the era of big data, the amount of data for optimization problems is constantly increasing, and the dimension is constantly increasing, and it has more complex multi-mode properties. Therefore, the traditional optimization algorithms can not meet the requirements of people. Inspired by the phenomenon that the state of matter changes tends to be stable,and according to universal gravitation and four particle motion modes, this paper proposes a novel Colon State Adaptive Optimization algorithm based on Gravitation(G-CSAO). In the function optimization experiment, the proposed algorithm has higher precision of optimization, and is not easy to fall into local optimal value, and can better cope with multi-peak problems and high-dimensional problems. The proposed model is combined with the minimum distance criterion in the pattern classification problem. The experimental results of UCI dataset show that the proposed model has better classification performance and application prospects.