Gated Graph Convolutional Recurrent Neural Networks. Ruiz, L., Gama, F., & Ribeiro, A. In 2019 27th European Signal Processing Conference (EUSIPCO), pages 1-5, Sep., 2019.
Paper doi abstract bibtex Graph processes model a number of important problems such as identifying the epicenter of an earthquake or predicting weather. In this paper, we propose a Graph Convolutional Recurrent Neural Network (GCRNN) architecture specifically tailored to deal with these problems. GCRNNs use convolutional filter banks to keep the number of trainable parameters independent of the size of the graph and of the time sequences considered. We also put forward Gated GCRNNs, a time-gated variation of GCRNNs akin to LSTMs. When compared with GNNs and another graph recurrent architecture in experiments using both synthetic and real-word data, GCRNNs significantly improve performance while using considerably less parameters.
@InProceedings{8902995,
author = {L. Ruiz and F. Gama and A. Ribeiro},
booktitle = {2019 27th European Signal Processing Conference (EUSIPCO)},
title = {Gated Graph Convolutional Recurrent Neural Networks},
year = {2019},
pages = {1-5},
abstract = {Graph processes model a number of important problems such as identifying the epicenter of an earthquake or predicting weather. In this paper, we propose a Graph Convolutional Recurrent Neural Network (GCRNN) architecture specifically tailored to deal with these problems. GCRNNs use convolutional filter banks to keep the number of trainable parameters independent of the size of the graph and of the time sequences considered. We also put forward Gated GCRNNs, a time-gated variation of GCRNNs akin to LSTMs. When compared with GNNs and another graph recurrent architecture in experiments using both synthetic and real-word data, GCRNNs significantly improve performance while using considerably less parameters.},
keywords = {channel bank filters;convolutional neural nets;graph theory;neural net architecture;recurrent neural nets;convolutional filter banks;earthquake;graph convolutional recurrent neural network architecture;epicenter;LSTMs;gated GCRNNs;Recurrent neural networks;Logic gates;Convolution;Transforms;Computer architecture;Convolutional neural networks;Data models;graph neural networks;recurrent neural networks;gating;graph processes},
doi = {10.23919/EUSIPCO.2019.8902995},
issn = {2076-1465},
month = {Sep.},
url = {https://www.eurasip.org/proceedings/eusipco/eusipco2019/proceedings/papers/1570533746.pdf},
}
Downloads: 0
{"_id":"9cDNzTd9RodATm8ZX","bibbaseid":"ruiz-gama-ribeiro-gatedgraphconvolutionalrecurrentneuralnetworks-2019","authorIDs":[],"author_short":["Ruiz, L.","Gama, F.","Ribeiro, A."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"firstnames":["L."],"propositions":[],"lastnames":["Ruiz"],"suffixes":[]},{"firstnames":["F."],"propositions":[],"lastnames":["Gama"],"suffixes":[]},{"firstnames":["A."],"propositions":[],"lastnames":["Ribeiro"],"suffixes":[]}],"booktitle":"2019 27th European Signal Processing Conference (EUSIPCO)","title":"Gated Graph Convolutional Recurrent Neural Networks","year":"2019","pages":"1-5","abstract":"Graph processes model a number of important problems such as identifying the epicenter of an earthquake or predicting weather. In this paper, we propose a Graph Convolutional Recurrent Neural Network (GCRNN) architecture specifically tailored to deal with these problems. GCRNNs use convolutional filter banks to keep the number of trainable parameters independent of the size of the graph and of the time sequences considered. We also put forward Gated GCRNNs, a time-gated variation of GCRNNs akin to LSTMs. When compared with GNNs and another graph recurrent architecture in experiments using both synthetic and real-word data, GCRNNs significantly improve performance while using considerably less parameters.","keywords":"channel bank filters;convolutional neural nets;graph theory;neural net architecture;recurrent neural nets;convolutional filter banks;earthquake;graph convolutional recurrent neural network architecture;epicenter;LSTMs;gated GCRNNs;Recurrent neural networks;Logic gates;Convolution;Transforms;Computer architecture;Convolutional neural networks;Data models;graph neural networks;recurrent neural networks;gating;graph processes","doi":"10.23919/EUSIPCO.2019.8902995","issn":"2076-1465","month":"Sep.","url":"https://www.eurasip.org/proceedings/eusipco/eusipco2019/proceedings/papers/1570533746.pdf","bibtex":"@InProceedings{8902995,\n author = {L. Ruiz and F. Gama and A. Ribeiro},\n booktitle = {2019 27th European Signal Processing Conference (EUSIPCO)},\n title = {Gated Graph Convolutional Recurrent Neural Networks},\n year = {2019},\n pages = {1-5},\n abstract = {Graph processes model a number of important problems such as identifying the epicenter of an earthquake or predicting weather. In this paper, we propose a Graph Convolutional Recurrent Neural Network (GCRNN) architecture specifically tailored to deal with these problems. GCRNNs use convolutional filter banks to keep the number of trainable parameters independent of the size of the graph and of the time sequences considered. We also put forward Gated GCRNNs, a time-gated variation of GCRNNs akin to LSTMs. When compared with GNNs and another graph recurrent architecture in experiments using both synthetic and real-word data, GCRNNs significantly improve performance while using considerably less parameters.},\n keywords = {channel bank filters;convolutional neural nets;graph theory;neural net architecture;recurrent neural nets;convolutional filter banks;earthquake;graph convolutional recurrent neural network architecture;epicenter;LSTMs;gated GCRNNs;Recurrent neural networks;Logic gates;Convolution;Transforms;Computer architecture;Convolutional neural networks;Data models;graph neural networks;recurrent neural networks;gating;graph processes},\n doi = {10.23919/EUSIPCO.2019.8902995},\n issn = {2076-1465},\n month = {Sep.},\n url = {https://www.eurasip.org/proceedings/eusipco/eusipco2019/proceedings/papers/1570533746.pdf},\n}\n\n","author_short":["Ruiz, L.","Gama, F.","Ribeiro, A."],"key":"8902995","id":"8902995","bibbaseid":"ruiz-gama-ribeiro-gatedgraphconvolutionalrecurrentneuralnetworks-2019","role":"author","urls":{"Paper":"https://www.eurasip.org/proceedings/eusipco/eusipco2019/proceedings/papers/1570533746.pdf"},"keyword":["channel bank filters;convolutional neural nets;graph theory;neural net architecture;recurrent neural nets;convolutional filter banks;earthquake;graph convolutional recurrent neural network architecture;epicenter;LSTMs;gated GCRNNs;Recurrent neural networks;Logic gates;Convolution;Transforms;Computer architecture;Convolutional neural networks;Data models;graph neural networks;recurrent neural networks;gating;graph processes"],"metadata":{"authorlinks":{}},"downloads":0},"bibtype":"inproceedings","biburl":"https://raw.githubusercontent.com/Roznn/EUSIPCO/main/eusipco2019url.bib","creationDate":"2021-02-11T19:15:22.084Z","downloads":0,"keywords":["channel bank filters;convolutional neural nets;graph theory;neural net architecture;recurrent neural nets;convolutional filter banks;earthquake;graph convolutional recurrent neural network architecture;epicenter;lstms;gated gcrnns;recurrent neural networks;logic gates;convolution;transforms;computer architecture;convolutional neural networks;data models;graph neural networks;recurrent neural networks;gating;graph processes"],"search_terms":["gated","graph","convolutional","recurrent","neural","networks","ruiz","gama","ribeiro"],"title":"Gated Graph Convolutional Recurrent Neural Networks","year":2019,"dataSources":["NqWTiMfRR56v86wRs","r6oz3cMyC99QfiuHW","KhfhF8P52iu5Szymq"]}