https://github.com/Malllabiisc/CompGCN

ICLR 2020: Composition-Based Multi-Relational Graph Convolutional Networks
https://github.com/Malllabiisc/CompGCN

deep-learning graph-convolutional-networks graph-representation-learning iclr2020 link-prediction pytorch relation-embeddings

Last synced: 3 months ago
JSON representation

ICLR 2020: Composition-Based Multi-Relational Graph Convolutional Networks

• Host: GitHub
• URL: https://github.com/Malllabiisc/CompGCN
• Owner: malllabiisc
• License: apache-2.0
• Created: 2020-01-04T12:03:00.000Z (almost 5 years ago)
• Default Branch: master
• Last Pushed: 2024-06-17T23:00:15.000Z (5 months ago)
• Last Synced: 2024-06-29T07:51:51.220Z (4 months ago)
• Topics: deep-learning, graph-convolutional-networks, graph-representation-learning, iclr2020, link-prediction, pytorch, relation-embeddings
• Language: Python
• Homepage:
• Size: 4.34 MB
• Stars: 584
• Watchers: 18
• Forks: 106
• Open Issues: 15
• Metadata Files:
  • Readme: Readme.md
  • License: LICENSE

Awesome Lists containing this project

• StarryDivineSky - Malllabiisc/CompGCN - GCN提出的Encoder-Decoder框架，在编码阶段将Entity Embedding和Realtion Embedding进行组合Aggregation，然后在解码阶段再采用类似TransE/H或者ConvE等方式对（h,r,t）三元组进行解码。因为它在编码阶段就引入了Realtion，使用同一套Realtion Embedding，使得表征学习更加精准。 (图嵌入、网络表征学习 / 网络服务_其他)