Modeling Relational Data with Graph Convolutional Networks thumbnail
Pause
Mute
Subtitles not available
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Modeling Relational Data with Graph Convolutional Networks

Published on Jul 10, 20182041 Views

Knowledge graphs enable a wide variety of applications, including question answering and information retrieval. Despite the great effort invested in their creation and maintenance, even the largest (e

Related categories

Chapter list

Modeling Relational Data with Graph Convolutional Networks00:00
Deep Learning for data on “grids” - 100:34
Deep Learning for data on “grids” - 201:55
Traditional vs. “deep” end-to-end learning02:14
Graph-structured / relational data - 102:59
Graph-structured / relational data - 203:13
Graph-structured / relational data - 303:30
Graph-structured / relational data - 403:36
Graph Convolutional Networks (GCNs)03:51
Recap: Convolutional neural networks (on grids)04:50
Graph convolutional networks (GCNs)06:08
Relational extension of GCNs08:59
How to scale?10:41
R-GCNs for entity classification and link prediction13:18
What do learned representations look like?14:54
Untitled15:34
Toy example (semi-supervised learning)15:52
R-GCNs for entity classification and link prediction16:27
Entity classification experiments18:12
Link prediction experiments - 119:41
Link prediction experiments - 221:16
Conclusions and future work22:14
Further reading23:28