摘要
该文提出了一种新的支持向量机学习算法—基于壳向量的增量学习算法(HVISVM)。选取一部分最有可能成为支持向量的样本—壳向量,再进行SVM增量学习。由于提取壳向量的过程只需线性规划运算,之后的训练过程又只需处理原训练样本中的一部分;增量学习既能继承先前所学习的知识,又能减少由于新样本的加入而重新学习的时间。使整个算法的训练速度大为提高。与经典支持向量机的快速算法比,精度相当,但速度可以提高数倍以上。
A new Geometric fast algorithm of support vector machines (HVISVM) and the concept of "hullvector" are proposed. The algorithm first extracts the set of hullvectors, which are most likely to be the support vector. Then the set of hullvectors are incremental trained as the new training samples to get the support vectors. The incremental train not only inherit knowledge which was studied before but also reduce the time of restudy when new samples are joined in. HVISVM reduces the time consumed by the QP problem in the SVM training in large degree, and highly speeds the whole training process of SVM. And this method does not decline the performance of SVM.
出处
《电子测量与仪器学报》
CSCD
2006年第6期94-97,共4页
Journal of Electronic Measurement and Instrumentation
关键词
模式识别
统计学习理论
支持向量机
壳向量
pattern recognition, statistical learning theory, support vector machine, hullvector.