Print Email Facebook Twitter Convolutional Neural Network Architectures for Signals Supported on Graphs Title Convolutional Neural Network Architectures for Signals Supported on Graphs Author Gama, F. (University of Pennsylvania) Marques, Antonio G. (Universidad Rey Juan Carlos) Leus, G.J.T. (TU Delft Signal Processing Systems) Ribeiro, Alejandro (University of Pennsylvania) Date 2019 Abstract Two architectures that generalize convolutional neural networks (CNNs) for the processing of signals supported on graphs are introduced. We start with the selection graph neural network (GNN), which replaces linear time invariant filters with linear shift invariant graph filters to generate convolutional features and reinterprets pooling as a possibly nonlinear subsampling stage where nearby nodes pool their information in a set of preselected sample nodes. A key component of the architecture is to remember the position of sampled nodes to permit computation of convolutional features at deeper layers. The second architecture, dubbed aggregation GNN, diffuses the signal through the graph and stores the sequence of diffused components observed by a designated node. This procedure effectively aggregates all components into a stream of information having temporal structure to which the convolution and pooling stages of regular CNNs can be applied. A multinode version of aggregation GNNs is further introduced for operation in large-scale graphs. An important property of selection and aggregation GNNs is that they reduce to conventional CNNs when particularized to time signals reinterpreted as graph signals in a circulant graph. Comparative numerical analyses are performed in a source localization application over synthetic and real-world networks. Performance is also evaluated for an authorship attribution problem and text category classification. Multinode aggregation GNNs are consistently the best-performing GNN architecture. Subject convolutional neural networksDeep learninggraph filtersgraph signal processingpooling To reference this document use: http://resolver.tudelft.nl/uuid:a7c0d871-4c79-4147-8335-885797b98fbf DOI https://doi.org/10.1109/TSP.2018.2887403 Embargo date 2019-08-15 ISSN 1053-587X Source IEEE Transactions on Signal Processing, 67 (4), 1034-1049 Bibliographical note Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public. Part of collection Institutional Repository Document type journal article Rights © 2019 F. Gama, Antonio G. Marques, G.J.T. Leus, Alejandro Ribeiro Files PDF Convolutional_Neural_Netw ... Graphs.pdf 3.49 MB Close viewer /islandora/object/uuid:a7c0d871-4c79-4147-8335-885797b98fbf/datastream/OBJ/view