摘要
针对使用径向基核函数的支持向量机,采用粒子群优化方法实现模型优化.基于训练集中样本之间的最近平均距离和最远平均距离,给出参数σ的取值空间,从而减小了超参数搜索的范围,并采用对数刻度进一步提高粒子群优化方法的参数搜索效率.与遗传算法和网格法的对比实验表明,所提出的方法收敛速度更快,得出的超参数更优.
For the radial basis function (RBF) kernel based support vector machines (SVM),particle swarm optimization (PSO) is employed to carry out the model optimization. The value space of the parameter σ is presented on the analysis of the mean shortest distance and mean furthest distance among the samples of the training set,thus the search region is reduced,and logarithmic scale is employed to further improve the search efficiency of PSO. Extensive experimental results on comparison with genetic algorithm and grid based approaches show that the proposed approach converges faster and produces better hyper-parameters.
出处
《控制与决策》
EI
CSCD
北大核心
2010年第3期367-370,377,共5页
Control and Decision
基金
国家自然科学基金项目(60975026)
陕西省自然科学研究计划项目(2007F19)
关键词
模型优化
支持向量机
粒子群优化
搜索效率
Model optimization
Support vector machine
Particle swarm optimization
Search efficiency