非均衡加权随机梯度下降SVM在线算法
发布时间：2018-08-04 07:46
【摘要】：随机梯度下降(stochastic gradient descent,SGD)方法已被应用于大规模支持向量机(support vector machine,SVM)训练,其在训练时采取随机选点的方式,对于非均衡分类问题,导致多数类点被抽取到的概率要远远大于少数类点,造成了计算上的不平衡。为了处理大规模非均衡数据分类问题,提出了加权随机梯度下降的SVM在线算法,对于多数类中的样例被赋予较小的权值,而少数类中的样例被赋予较大的权值,然后利用加权随机梯度下降算法对SVM原问题进行求解,减少了超平面向少数类的偏移,较好地解决了大规模学习中非均衡数据的分类问题。
[Abstract]:Stochastic gradient descent (stochastic gradient descenting (stochastic gradient) method has been applied to large-scale support vector machine (support vector machine) training. The probability of extracting most of the points is much higher than that of a few, which results in the imbalance of calculation. In order to deal with the problem of large-scale disequilibrium data classification, a weighted stochastic gradient descent SVM online algorithm is proposed, in which the sample in most classes is given a smaller weight, while the sample in a few classes is given a larger weight. Then the weighted stochastic gradient descent algorithm is used to solve the original SVM problem, which reduces the deviation of the superplane for a few classes and solves the classification problem of unbalanced data in large-scale learning.
【作者单位】： 河北大学数学与信息科学学院河北省机器学习与计算智能重点实验室;
【基金】：河北省自然科学基金No.F2015201185~~
【分类号】：TP181
本文编号：2163139
[Abstract]:Stochastic gradient descent (stochastic gradient descenting (stochastic gradient) method has been applied to large-scale support vector machine (support vector machine) training. The probability of extracting most of the points is much higher than that of a few, which results in the imbalance of calculation. In order to deal with the problem of large-scale disequilibrium data classification, a weighted stochastic gradient descent SVM online algorithm is proposed, in which the sample in most classes is given a smaller weight, while the sample in a few classes is given a larger weight. Then the weighted stochastic gradient descent algorithm is used to solve the original SVM problem, which reduces the deviation of the superplane for a few classes and solves the classification problem of unbalanced data in large-scale learning.
【作者单位】： 河北大学数学与信息科学学院河北省机器学习与计算智能重点实验室;
【基金】：河北省自然科学基金No.F2015201185~~
【分类号】：TP181
【相似文献】
相关期刊论文 前3条
1 谢汉中;在PLA里交叉点故障定位的在线算法[J];江苏电机工程;1987年01期
2 帅典勋;在可编程序的逻辑阵列(PLA)中交叉点故障定位的一种在线算法[J];计算机工程;1984年06期
3 贺文武;;在线核学习的一般形式探讨[J];福建工程学院学报;2010年04期
相关会议论文 前2条
1 何宇新;韩志刚;;多层递阶方法时变参数估值最佳初值和最佳跟踪的在线算法[A];1992年中国控制与决策学术年会论文集[C];1992年
2 尹焕平;孙宗海;;基于自然梯度的支持向量回归在线算法[A];2009中国控制与决策会议论文集（3）[C];2009年
相关博士学位论文 前1条
1 黄禹潇;面向复杂诊断贝叶斯网实时推理问题的离线、在线算法的研究[D];吉林大学;2012年
相关硕士学位论文 前1条
1 王明岳;m射线路径上移动目标搜索的在线算法研究[D];复旦大学;2009年
，本文编号：2163139
本文链接：https://www.wllwen.com/kejilunwen/zidonghuakongzhilunwen/2163139.html
