Deep learning via semi-supervised embedding

Abstract

We show how nonlinear embedding algorithms popular for use with <i>shallow</i> semi-supervised learning techniques such as kernel methods can be applied to deep multilayer architectures, either as a regularizer at the output layer, or on each layer of the architecture. This provides a simple alternative to existing approaches to <i>deep</i> learning whilst yielding competitive error rates compared to those methods, and existing <i>shallow</i> semi-supervised techniques.

DOI: 10.1145/1390156.1390303
View Slides

Extracted Key Phrases

0502008200920102011201220132014201520162017
Citations per Year

401 Citations

Semantic Scholar estimates that this publication has 401 citations based on the available data.

See our FAQ for additional information.

Cite this paper

@inproceedings{Weston2008DeepLV, title={Deep learning via semi-supervised embedding}, author={Jason Weston and Fr{\'e}d{\'e}ric Ratle and Ronan Collobert}, booktitle={ICML}, year={2008} }