Using SVMs with randomised feature spaces: an extreme learning approach
Using SVMs with randomised feature spaces: an extreme learning approach
复制标题
DOI：
--
复制
发表时间：
2010
期刊：
影响因子：
--
通讯作者：
Benoît Frénay;M. Verleysen
中科院分区：
文献类型：
--
作者：
Benoît Frénay;M. Verleysen
极限学习机是快速的模型，在准确性方面几乎可与标准支持向量机（SVM）相媲美，但速度要快得多。然而，它们优化的是误差平方和，而支持向量机是最大间隔分类器。本文提议通过定义一种新的核来融合这两种方法。这个核由极限学习机的第一层计算得出，并用于训练支持向量机。实验表明，这种新核在准确性方面可与标准的径向基函数（RBF）核相比较，且速度更快。实际上，实验表明，随机核背后的极限学习机的神经元数量无需调整，并且可以设置为一个足够的值，而不会显著改变准确性。
Extreme learning machines are fast models which almost compare to standard SVMs in terms of accuracy, but are much faster. However, they optimise a sum of squared errors whereas SVMs are maximum-margin classifiers. This paper proposes to merge both approaches by defining a new kernel. This kernel is computed by the first layer of an extreme learning machine and used to train a SVM. Experiments show that this new kernel compares to the standard RBF kernel in terms of accuracy and is faster. Indeed, experiments show that the number of neurons of the ELM behind the randomised kernel does not need to be tuned and can be set to a sufficient value without altering the accuracy significantly.