arXiv (Cornell University)· 2016· Preprint
Semi-Supervised Classification with Graph Convolutional Networks
- 8,069citations
- 2016year
Short summary
A new scalable graph convolutional network (GCN) model achieves state-of-the-art semi-supervised learning on graph-structured data, outperforming prior methods on citation and knowledge graph datasets.
AI-generated from the title and abstract; the full text is not read.
Key points
- Presents a scalable graph convolutional network (GCN) for semi-supervised learning on graph data.
- Model uses a localized first-order approximation of spectral graph convolutions.
- Achieves linear scaling with the number of graph edges.
- Learns node representations that capture local graph structure and node features.
- Outperforms related methods on citation and knowledge graph datasets.
AI-generated from the title and abstract; the full text is not read.
Abstract
We present a scalable approach for semi-supervised learning on graph-structured data that is based on an efficient variant of convolutional neural networks which operate directly on graphs. We motivate the choice of our convolutional architecture via a localized first-order approximation of spectral graph convolutions. Our model scales linearly in the number of graph edges and learns hidden layer representations that encode both local graph structure and features of nodes. In a number of experiments on citation networks and on a knowledge graph dataset we demonstrate that our approach outperforms related methods by a significant margin.
The authors' abstract, as published at the source. arXiv (Cornell University), 2016 · DOI ↗
Continue with a free account
Ask the paper: 3 free questions a day about this paper; save it, get its citation, new summaries every day for your field. Takeaways are Premium.
Continue free on the webSign in with Google or Apple; no card needed. You come back to this paper.
On your phone:
Field: Artificial Intelligence
Artificial IntelligenceComputer Science