Gated Graph Recurrent Neural Networks

Graph processes exhibit a temporal structure determined by the sequence index and and a spatial structure determined by the graph support. To learn from graph processes, an information processing architecture must then be able to exploit both underlying structures. We introduce Graph Recurrent Neura...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on signal processing 2020, Vol.68, p.6303-6318
Main Authors: Ruiz, Luana, Gama, Fernando, Ribeiro, Alejandro
Format: Article
Language:English
Subjects:
Convolution
Data processing
gating
graph convolutions
Graph recurrent neural networks
graph signal processing
Indexes
Logic gates
Neural networks
Numerical stability
Permutations
Recurrent neural networks
Signal processing
stability
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Graph processes exhibit a temporal structure determined by the sequence index and and a spatial structure determined by the graph support. To learn from graph processes, an information processing architecture must then be able to exploit both underlying structures. We introduce Graph Recurrent Neural Networks (GRNNs) as a general learning framework that achieves this goal by leveraging the notion of a recurrent hidden state together with graph signal processing (GSP). In the GRNN, the number of learnable parameters is independent of the length of the sequence and of the size of the graph, guaranteeing scalability. We prove that GRNNs are permutation equivariant and that they are stable to perturbations of the underlying graph support. To address the problem of vanishing gradients, we also put forward gated GRNNs with three different gating mechanisms: time, node and edge gates. In numerical experiments involving both synthetic and real datasets, time-gated GRNNs are shown to improve upon GRNNs in problems with long term dependencies, while node and edge gates help encode long range dependencies present in the graph. The numerical results also show that GRNNs outperform GNNs and RNNs, highlighting the importance of taking both the temporal and graph structures of a graph process into account.
ISSN:1053-587X
1941-0476
DOI:10.1109/TSP.2020.3033962