ePrints@IIScePrints@IISc Home | About | Browse | Latest Additions | Advanced Search | Contact | Help

Variational graph autoencoders for multiview canonical correlation analysis

Kaloga, Y and Borgnat, P and Chepuri, SP and Abry, P and Habrard, A (2021) Variational graph autoencoders for multiview canonical correlation analysis. In: Signal Processing, 188 .

[img] PDF
sig_pro_188_2021.pdf - Published Version
Restricted to Registered users only

Download (2MB) | Request a copy
Official URL: https://doi.org/10.1016/j.sigpro.2021.108182


We present a novel approach for multiview canonical correlation analysis based on a variational graph neural network model. We propose a nonlinear model which takes into account the available graph-based geometric constraints while being scalable to large-scale datasets with multiple views. This model combines the probabilistic interpretation of CCA with an autoencoder architecture based on graph convolutional neural network layers. Experiments with the proposed method are conducted on classification, clustering, and recommendation tasks on real datasets. The algorithm is competitive with state-of-the-art multiview representation learning techniques, in addition to being scalable and robust to instances with missing views. © 2021 Elsevier B.V.

Item Type: Journal Article
Publication: Signal Processing
Publisher: Elsevier B.V.
Additional Information: The copyright for this article belongs to Elsevier B.V.
Keywords: Classification (of information); Correlation methods; Large dataset; Learning systems; Multilayer neural networks; Network layers, Autoencoders; Canonical correlations analysis; Dimensionality reduction; Graph neural networks; Graph neurals network; Multi-views; Multiview representation learning; Neural network modelling; Non-linear model; Variational inference, Graphic methods
Department/Centre: Division of Electrical Sciences > Electrical Communication Engineering
Date Deposited: 24 Sep 2021 07:57
Last Modified: 24 Sep 2021 07:57
URI: http://eprints.iisc.ac.in/id/eprint/69742

Actions (login required)

View Item View Item