L2-norm multiple kernel learning and its application to biomedical data fusion.
L2-norm multiple kernel learning and its application to biomedical data fusion.
复制标题
DOI：
10.1186/1471-2105-11-309
复制
发表时间：
2010-06-08
影响因子：
3
通讯作者：
Moreau Y
中科院分区：
文献类型：
--
作者：
Yu S;Falck T;Daemen A;Tranchevent LC;Suykens JA;De Moor B;Moreau Y
本文在多核支持向量机的对偶问题中引入了优化不同范数的概念。范数的选择产生了多核学习的不同扩展，如L∞、L1和L2多核学习。特别是，L2MKL是一种导致非稀疏最优核系数的新方法，它不同于现有的L∞MKL方法优化的稀疏核系数。在实际的生物医学应用中，与稀疏集成方法相比，L2 MKL在彻底组合异质数据源中的互补信息方面具有更多的优势。从理论上分析了对偶问题中核的L2优化与原问题中的L2系数正则化之间的关系。理解对偶L2问题可以为MKL提供一个统一的观点，并使我们能够将L2方法扩展到广泛的机器学习问题。我们实现了L2MKL用于排序和分类问题，并与稀疏的L∞和平均的L1MKL方法的性能进行了比较。实验在六个真实的生物医学数据集和两个大规模的UCI数据集上进行。L2 MKL在大多数基准数据集上产生了更好的性能。特别是，我们提出了一种新的L2MKL最小二乘支持向量机(LSSVM)算法，该算法被证明是一种适用于大规模数据集处理的高效且有前途的分类器。本文扩展了基于MKL的基因组数据融合的统计框架。在我们认为大多数数据源与当前问题相关并希望避免L∞MKL中出现的“赢家通吃”效应的情况下，允许数据源上的非稀疏权重是一个有吸引力的选择，这可能会对前瞻性研究的性能造成不利影响。优化L2内核的概念可以直接扩展到排名、分类、回归和集群算法。针对MKL的计算负担问题，提出了几种基于最小二乘支持向量机的MKL算法。在真实数据集上的系统比较表明，最小二乘支持向量机MKL算法具有与传统支持向量机MKL算法相当的性能。大规模数值实验表明，当模型转化为半无限规划时，最小二乘支持向量机MKL比支持向量机MKL更有效。本文实现的算法的matlab代码可以从http://homes.esat.kuleuven.be/~sistawww/bioi/syu/l2lssvm.html.下载
This paper introduces the notion of optimizing different norms in the dual problem of support vector machines with multiple kernels. The selection of norms yields different extensions of multiple kernel learning (MKL) such as L∞, L1, and L2 MKL. In particular, L2 MKL is a novel method that leads to non-sparse optimal kernel coefficients, which is different from the sparse kernel coefficients optimized by the existing L∞ MKL method. In real biomedical applications, L2 MKL may have more advantages over sparse integration method for thoroughly combining complementary information in heterogeneous data sources. We provide a theoretical analysis of the relationship between the L2 optimization of kernels in the dual problem with the L2 coefficient regularization in the primal problem. Understanding the dual L2 problem grants a unified view on MKL and enables us to extend the L2 method to a wide range of machine learning problems. We implement L2 MKL for ranking and classification problems and compare its performance with the sparse L∞ and the averaging L1 MKL methods. The experiments are carried out on six real biomedical data sets and two large scale UCI data sets. L2 MKL yields better performance on most of the benchmark data sets. In particular, we propose a novel L2 MKL least squares support vector machine (LSSVM) algorithm, which is shown to be an efficient and promising classifier for large scale data sets processing. This paper extends the statistical framework of genomic data fusion based on MKL. Allowing non-sparse weights on the data sources is an attractive option in settings where we believe most data sources to be relevant to the problem at hand and want to avoid a "winner-takes-all" effect seen in L∞ MKL, which can be detrimental to the performance in prospective studies. The notion of optimizing L2 kernels can be straightforwardly extended to ranking, classification, regression, and clustering algorithms. To tackle the computational burden of MKL, this paper proposes several novel LSSVM based MKL algorithms. Systematic comparison on real data sets shows that LSSVM MKL has comparable performance as the conventional SVM MKL algorithms. Moreover, large scale numerical experiments indicate that when cast as semi-infinite programming, LSSVM MKL can be solved more efficiently than SVM MKL. The MATLAB code of algorithms implemented in this paper is downloadable from http://homes.esat.kuleuven.be/~sistawww/bioi/syu/l2lssvm.html.
登录
查看更多内容
影响因子：
5.8
作者：
Lanckriet, GRG;De Bie, T;Noble, WS
通讯作者：
Noble, WS
影响因子：
12.3
作者：
Daemen, Anneleen;Gevaert, Olivier;De Moor, Bart
通讯作者：
De Moor, Bart
DOI：
10.1007/978-1-84800-155-8_7
发表时间：
2008-01-01
期刊：
RECENT ADVANCES IN LEARNING AND CONTROL
影响因子：
--
作者：
Grant, Michael C.;Boyd, Stephen P.
通讯作者：
Boyd, Stephen P.
DOI：
10.1016/0377-0427(92)00122-p
发表时间：
1994-07-29
影响因子：
2.4
作者：
REEMTSEN, R
通讯作者：
REEMTSEN, R
影响因子：
30.8
作者：
Eeles, Rosalind A.;Kote-Jarai, Zsofia;Easton, Douglas F.
通讯作者：
Easton, Douglas F.