Transcription of Graph Transformer Networks - NeurIPS
{{id}} {{{paragraph}}}
Example: confidence
Graph Transformer Networks - NeurIPS
Graph Transformer NetworksSeongjun Yun, Minbyul Jeong, Raehyun Kim, Jaewoo Kang , Hyunwoo J. Kim Department of Computer Science and EngineeringKorea University{ysj5419, minbyuljeong, raehyun, kangj, neural Networks (GNNs) have been widely used in representation learning ongraphs and achieved state-of-the-art performance in tasks such as node classificationand link prediction. However, most existing GNNs are designed to learn noderepresentations on thefixedandhomogeneousgraphs. The limitations especiallybecome problematic when learning representations on a misspecified Graph oraheterogeneousgraph that consists of various types of nodes and edges. Inthis paper, we propose Graph Transformer Networks (GTNs) that are capable ofgenerating new Graph structures, which involve identifying useful connectionsbetween unconnected nodes on the original Graph , while learning effective noderepresentation on the new graphs in an end-to-end fashion.}
GTNs can be viewed as a graph analogue of Spatial Transformer Networks [16] which explicitly learn spatial transformations of input images or features. The main challenge to ... meta-paths that are the composite relations of multiple edge types, as a preprocessing, it converts the
Loading..
Tags:
Information
Domain:
Source:
Link to this page:
Please notify us if you found a problem with this document: