About cookies on this site Our websites require some cookies to function properly (required). In addition, other cookies may be used with your consent to analyze site usage, improve the user experience and for advertising. For more information, please review your options. By visiting our website, you agree to our processing of information as described in IBM’sprivacy statement. To provide a smooth navigation, your cookie preferences will be shared across the IBM web domains listed here.
Publication
NeurIPS 2019
Workshop paper
Tensor graph neural networks for learning on time varying graphs
Abstract
Many irregular domains such as social networks, financial transactions, neuronconnections, and natural language structures are represented as graphs. A varietyof graph neural networks (GNNs) have been successfully applied for representa-tion learning and prediction on such graphs. However, in many of the applica-tions, the underlying graph changes over time and existing GNNs are inadequatefor handling such time varying graphs. In this paper we propose a novel techniquefor learning embeddings of time varying graphs based on a tensor framework. Themethod extends the popular graph convolutional network (GCN) for learning rep-resentations of time varying graphs using the recently proposed tensor M-producttechnique. Numerical experiments on four real datasets demonstrate that our pro-posed method outperforms a baseline method when used for edge classification.