You are here

Learning Multiple Views with Orthogonal Denoising Autoencoders

Authors: 

TengQi Ye, Tianchun Wang, Kevin McGuinness, Yu Guo, Cathal Gurrin

Publication Type: 
Refereed Conference Meeting Proceeding
Abstract: 
Multi-view learning techniques are necessary when data is described by multiple distinct feature sets because single-view learning algorithms tend to overfit on these high-dimensional data. Prior success- ful approaches followed either consensus or complementary principles. Recent work has focused on learning both the shared and private latent spaces of views in order to take advantage of both principles. However, these methods can not ensure that the latent spaces are strictly indepen- dent through encouraging the orthogonality in their objective functions. Also little work has explored representation learning techniques for multi- view learning. In this paper, we use the denoising autoencoder to learn shared and private latent spaces, with orthogonal constraints — discon- necting every private latent space from the remaining views. Instead of computationally expensive optimization, we adapt the backpropagation algorithm to train our model.
Conference Name: 
Multimedia Modelling
Proceedings: 
Proceedings of Multimedia Modelling, Miami, Fl, USA
Digital Object Identifer (DOI): 
10.1007/978-3-319-27671-7_26
Publication Date: 
04/01/2016
Conference Location: 
United States of America
Research Group: 
Institution: 
Dublin City University (DCU)
Open access repository: 
Yes