基于支持向量机泛化误差界的多核学习方法

来源 :2011年全国理论计算机科学学术年会 | 被引量 : 0次 | 上传用户:iam156
下载到本地 , 更方便阅读
声明 : 本文档内容版权归属内容提供方 , 如果您对本文有版权争议 , 可与客服联系进行内容授权或下架
论文部分内容阅读
支持向量机(Support Vector Machine,SVM)是当前机器学习、模式识别和数据挖掘等领域的重要学习方法,核函数选择是研究和应用SVM的关键.传统模型选择方法利用数据从给定的候选集中选择单一核函数,近来的理论和应用表明:利用多核代替单核不仅能增加SVM的灵活性,而且能增强分类器的性能和可解释性.
其他文献
会议
会议
会议
会议
会议
In this presentation,the insight mechanism of safety is reviewed.The safety of metal oxide cathode based vehicular batteries is improved and new understanding
会议
Diamond has the highest surface acoustic wave (SAW) velocity among all materials.It can also provide high breakdown voltage,radiation hardness and high thermal
The Cr and Ti based boron nitride thin films,including Cr–B–N and Ti-Cr–B–N coatings with various boron contents were deposited by a co-sputtering process u
会议
A new micro-thermoelectric power generator module with high packing density of film thermoelectric legs has been proposed,in which a larger number of p-type and