Multiview Variational Graph Autoencoders for Canonical Correlation Analysis
This work addresses multiview representation learning for researchers and practitioners needing scalable nonlinear methods with graph constraints.
The authors tackled the problem of multiview canonical correlation analysis by developing a nonlinear variational model that incorporates graph-based geometric constraints while maintaining scalability for large datasets. Their approach achieved competitive performance with state-of-the-art methods on classification, clustering, and recommendation tasks using real datasets.
We present a novel multiview canonical correlation analysis model based on a variational approach. This is the first nonlinear model that takes into account the available graph-based geometric constraints while being scalable for processing large scale datasets with multiple views. It is based on an autoencoder architecture with graph convolutional neural network layers. We experiment with our approach on classification, clustering, and recommendation tasks on real datasets. The algorithm is competitive with state-of-the-art multiview representation learning techniques.