Hypergraph convolution and hypergraph attention
Recently, graph neural networks have attracted great attention and achieved prominent performance in various research fields. Most of those algorithms have assumed pairwise relationships of objects of interest. However, in many real applications, the relationships between objects are in higher-order...
| Main Authors: | , , |
|---|---|
| Format: | Journal article |
| Language: | English |
| Published: |
Elseveir
2020
|
| _version_ | 1826261201384898560 |
|---|---|
| author | Bai, S Zhang, F Torr, PHS |
| author_facet | Bai, S Zhang, F Torr, PHS |
| author_sort | Bai, S |
| collection | OXFORD |
| description | Recently, graph neural networks have attracted great attention and achieved prominent performance in various research fields. Most of those algorithms have assumed pairwise relationships of objects of interest. However, in many real applications, the relationships between objects are in higher-order, beyond a pairwise formulation. To efficiently learn deep embeddings on the high-order graph-structured data, we introduce two end-to-end trainable operators to the family of graph neural networks, i.e., hypergraph convolution and hypergraph attention. Whilst hypergraph convolution defines the basic formulation of performing convolution on a hypergraph, hypergraph attention further enhances the capacity of representation learning by leveraging an attention module. With the two operators, a graph neural network is readily extended to a more flexible model and applied to diverse applications where non-pairwise relationships are observed. Extensive experimental results with semi-supervised node classification demonstrate the effectiveness of hypergraph convolution and hypergraph attention. |
| first_indexed | 2024-03-06T19:17:48Z |
| format | Journal article |
| id | oxford-uuid:1904e878-a7c1-4bd3-be1e-701049091d19 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-06T19:17:48Z |
| publishDate | 2020 |
| publisher | Elseveir |
| record_format | dspace |
| spelling | oxford-uuid:1904e878-a7c1-4bd3-be1e-701049091d192022-03-26T10:46:31ZHypergraph convolution and hypergraph attentionJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:1904e878-a7c1-4bd3-be1e-701049091d19EnglishSymplectic ElementsElseveir2020Bai, SZhang, FTorr, PHSRecently, graph neural networks have attracted great attention and achieved prominent performance in various research fields. Most of those algorithms have assumed pairwise relationships of objects of interest. However, in many real applications, the relationships between objects are in higher-order, beyond a pairwise formulation. To efficiently learn deep embeddings on the high-order graph-structured data, we introduce two end-to-end trainable operators to the family of graph neural networks, i.e., hypergraph convolution and hypergraph attention. Whilst hypergraph convolution defines the basic formulation of performing convolution on a hypergraph, hypergraph attention further enhances the capacity of representation learning by leveraging an attention module. With the two operators, a graph neural network is readily extended to a more flexible model and applied to diverse applications where non-pairwise relationships are observed. Extensive experimental results with semi-supervised node classification demonstrate the effectiveness of hypergraph convolution and hypergraph attention. |
| spellingShingle | Bai, S Zhang, F Torr, PHS Hypergraph convolution and hypergraph attention |
| title | Hypergraph convolution and hypergraph attention |
| title_full | Hypergraph convolution and hypergraph attention |
| title_fullStr | Hypergraph convolution and hypergraph attention |
| title_full_unstemmed | Hypergraph convolution and hypergraph attention |
| title_short | Hypergraph convolution and hypergraph attention |
| title_sort | hypergraph convolution and hypergraph attention |
| work_keys_str_mv | AT bais hypergraphconvolutionandhypergraphattention AT zhangf hypergraphconvolutionandhypergraphattention AT torrphs hypergraphconvolutionandhypergraphattention |