Convolutional Neural Network Architectures for Signals Supported on Graphs

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.

arXiv commentSubmitted to IEEE Transactions on Signal Processing

References in corpus (17)

• Adam: A Method for Stochastic Optimization
• Efficient Estimation of Word Representations in Vector Space
• Semi-Supervised Classification with Graph Convolutional Networks
• The Emerging Field of Signal Processing on Graphs: Extending High-Dimensional Data Analysis to Networks and Other Irregular Domains
• Discrete Signal Processing on Graphs
• Deep Convolutional Networks on Graph-Structured Data
• Asymptotic analysis of the stochastic block model for modular networks and its algorithmic applications
• Discrete Signal Processing on Graphs: Sampling Theory
• Diffusion-Convolutional Neural Networks
• Efficient Sampling Set Selection for Bandlimited Graph Signals Using Graph Spectral Proxies
• Signals on Graphs: Uncertainty Principle and Sampling
• Sampling of graph signals with successive local aggregations
• Topology Adaptive Graph Convolutional Networks
• A Comprehensive Survey of Graph Embedding: Problems, Techniques and Applications
• Authorship Attribution through Function Word Adjacency Networks
• Convolutional neural networks on irregular domains based on approximate vertex-domain translations
• CNN as Guided Multi-layer RECOS Transform

Cited by in corpus (64)

• Kernel Mean Embedding of Distributions: A Review and Beyond
• Learning Decentralized Controllers for Robot Swarms with Graph Neural Networks
• Optimal Wireless Resource Allocation with Random Edge Graph Neural Networks
• Fault Location in Power Distribution Systems via Deep Graph Convolutional Networks
• Stability Properties of Graph Neural Networks
• Gated Graph Recurrent Neural Networks
• Missing Data Imputation with Adversarially-trained Graph Convolutional Networks
• Graphs, Convolutions, and Neural Networks: From Graph Filters to Graph Neural Networks
• Signal Processing on Higher-Order Networks: Livin' on the Edge ... and Beyond
• SR2CNN: Zero-Shot Learning for Signal Recognition
• Advances in Distributed Graph Filtering
• Graph Unrolling Networks: Interpretable Neural Networks for Graph Signal Denoising
• Learning Decentralized Wireless Resource Allocations with Graph Neural Networks
• Deep Unsupervised Learning of 3D Point Clouds via Graph Topology Inference and Filtering
• Hierarchical Representation Learning in Graph Neural Networks with Node Decimation Pooling
• Invariance-Preserving Localized Activation Functions for Graph Neural Networks
• Link Scheduling using Graph Neural Networks
• Coordinated Sum-Rate Maximization in Multicell MU-MIMO with Deep Unrolling
• Graphon Signal Processing
• Graphon Neural Networks and the Transferability of Graph Neural Networks
• Stability of Graph Scattering Transforms
• Tensor Graph Convolutional Networks for Multi-relational and Robust Learning
• Stochastic Graph Neural Networks
• Graph Neural Networks for Decentralized Multi-Agent Perimeter Defense
• Algebraic Neural Networks: Stability to Deformations
• Wide and Deep Graph Neural Network with Distributed Online Learning
• Graph-based Algorithm Unfolding for Energy-aware Power Allocation in Wireless Networks
• The Emergence of Adversarial Communication in Multi-Agent Reinforcement Learning
• Graph Policy Gradients for Large Scale Robot Control
• Convolutional Learning on Multigraphs
• Graph Neural Networks for Motion Planning
• Message-Aware Graph Attention Networks for Large-Scale Multi-Robot Path Planning
• Training Robust Graph Neural Networks with Topology Adaptive Edge Dropping
• An Underparametrized Deep Decoder Architecture for Graph Signals
• Graph Neural Networks for Decentralized Multi-Robot Submodular Action Selection
• Distributed Scheduling using Graph Neural Networks
• Graph and graphon neural network stability
• Unsupervised Learning for Asynchronous Resource Allocation in Ad-hoc Wireless Networks
• Distributionally Robust Semi-Supervised Learning Over Graphs
• Sampling and Recovery of Graph Signals based on Graph Neural Networks
• Stability of Neural Networks on Riemannian Manifolds
• Signal Processing on Directed Graphs
• Graph GOSPA metric: a metric to measure the discrepancy between graphs of different sizes
• VGAI: End-to-End Learning of Vision-Based Decentralized Controllers for Robot Swarms
• Convolutional Filtering and Neural Networks with Non Commutative Algebras
• Scalable Perception-Action-Communication Loops with Convolutional and Graph Neural Networks
• Attentional Graph Convolutional Networks for Knowledge Concept Recommendation in MOOCs in a Heterogeneous View
• Stability of Graph Convolutional Neural Networks to Stochastic Perturbations
• Reliable Graph Neural Network Explanations Through Adversarial Training
• Graph learning under spectral sparsity constraints
• Quiver Signal Processing (QSP)
• Towards an Efficient and General Framework of Robust Training for Graph Neural Networks
• Spherical Convolutional Neural Networks: Stability to Perturbations in SO(3)
• Wide and Deep Graph Neural Networks with Distributed Online Learning
• Deep Constraint-based Propagation in Graph Neural Networks
• Blind Demixing of Diffused Graph Signals
• Graph Neural Networks for Distributed Linear-Quadratic Control
• Online Distributed Learning over Graphs with Multitask Graph-Filter Models
• PanRep: Graph neural networks for extracting universal node embeddings in heterogeneous graphs
• Untrained Graph Neural Networks for Denoising
• Graph Neural Network for Large-Scale Network Localization
• Enhancing Geometric Deep Learning via Graph Filter Deconvolution
• Robust, Deep, and Reinforcement Learning for Management of Communication and Power Networks
• Resource Allocation via Graph Neural Networks in Free Space Optical Fronthaul Networks