A deep quasi-linear kernel composition method for support vector machines

Weite Li, Takayuki Furuzuki, Benhui Chen

研究成果: Conference contribution

2 引用 (Scopus)

抜粋

In this paper, we introduce a data-dependent kernel called deep quasi-linear kernel, which can directly gain a profit from a pre-trained feedforward deep network. Firstly, a multi-layer gated bilinear classifier is formulated to mimic the functionality of a feed-forward neural network. The only difference between them is that the activation values of hidden units in the multi-layer gated bilinear classifier are dependent on a pre-trained neural network rather than a pre-defined activation function. Secondly, we demonstrate the equivalence between the multi-layer gated bilinear classifier and an SVM with a deep quasi-linear kernel. By deriving a kernel composition function, traditional optimization algorithms for a kernel SVM can be directly implemented to implicitly optimize the parameters of the multi-layer gated bilinear classifier. Experimental results on different data sets show that our proposed classifier obtains an ability to outperform both an SVM with a RBF kernel and the pre-trained feedforward deep network.

元の言語English
ホスト出版物のタイトル2016 International Joint Conference on Neural Networks, IJCNN 2016
出版者Institute of Electrical and Electronics Engineers Inc.
ページ1639-1645
ページ数7
2016-October
ISBN(電子版)9781509006199
DOI
出版物ステータスPublished - 2016 10 31
イベント2016 International Joint Conference on Neural Networks, IJCNN 2016 - Vancouver, Canada
継続期間: 2016 7 242016 7 29

Other

Other2016 International Joint Conference on Neural Networks, IJCNN 2016
Canada
Vancouver
期間16/7/2416/7/29

    フィンガープリント

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence

これを引用

Li, W., Furuzuki, T., & Chen, B. (2016). A deep quasi-linear kernel composition method for support vector machines. : 2016 International Joint Conference on Neural Networks, IJCNN 2016 (巻 2016-October, pp. 1639-1645). [7727394] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IJCNN.2016.7727394