Deep learning via semi-supervised embedding

被引：0

作者：

Weston, Jason ^{[1
]}

Ratle, Frédéric ^{[2
]}

Mobahi, Hossein ^{[3
]}

Collobert, Ronan ^{[4
]}

机构：

[1] Google, NY, United States

[2] Nuance Communications, Montreal, Canada

[3] Department of Computer Science, University of Illinois Urbana-Champaign, United States

[4] IDIAP Research Institute, Martigny, Switzerland

来源：

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | 2012年 / 7700 LECTURE NO卷

关键词：

Neural networks;

D O I：

10.1007/978-3-642-35289-8-34

中图分类号：

TB18 [人体工程学]; Q98 [人类学];

学科分类号：

030303 ; 1201 ;

摘要：

We show how nonlinear semi-supervised embedding algorithms popular for use with shallow learning techniques such as kernel methods can be easily applied to deep multi-layer architectures, either as a regularizer at the output layer, or on each layer of the architecture. Compared to standard supervised backpropagation this can give significant gains. This trick provides a simple alternative to existing approaches to semi-supervised deep learning whilst yielding competitive error rates compared to those methods, and existing shallow semi-supervised techniques. © Springer-Verlag Berlin Heidelberg 2012.

引用

页码：639 / 655