0.25
0.5
0.75
1.25
1.5
1.75
2
Deep Learning on Graphs
Published on Jan 17, 20192111 Views
Machine learning on graphs is an important and ubiquitous task with applications ranging from drug design to friendship recommendation in social networks. The primary challenge in this domain is findi
Related categories
Chapter list
Deep Learning in Graphs00:00
Networks: Common Language00:39
Tasks on Networks01:00
Example: Node Classification01:46
Machine Learning Lifecycle02:33
Deep Learning in Graphs03:35
Example of an Embedding04:06
Why Learn Embeddings?04:54
Embedding Nodes05:35
Why is it Hard? - 1 06:40
Why is it Hard? - 207:15
GraphSAGE: Graph Neural Networks08:27
Untitled08:39
From Images to Networks09:17
Real-World Graphs09:48
A Naïve Approach10:08
Untitled11:14
Our Approach: GraphSAGE - 111:58
Our Approach: GraphSAGE - 212:46
Our Approach: GraphSAGE - 314:02
Inductive Capability - 217:48
Inductive Capability - 118:40
Inductive Capability - 319:40
GraphSAGE: Training (1)20:15
GraphSAGE: Training (2)21:48
GraphSAGE: Benefits22:58
PinSAGE for Recommender Systems26:01
Application: Pinterest27:05
Task Overview27:58
Curriculum Learning29:06
GraphSAGE Training31:16
GraphSAGE: Inference32:37
Experiments33:30
PinSAGE Performance34:54
PinSAGE Recommendations - 135:38
PinSAGE Recommendations - 236:36
Decagon: Drug Side Effect Prediction37:40
Polypharmacy side effects38:05
Graph: Molecular, Drug & Patient39:07
Decagon: Graph Neural Net40:46
Encoder: Embeddings - 141:17
Encoder: Embeddings - 242:44
Decoder: Link Prediction43:16
Results: Side Effect Prediction44:10
De novo Predictions - 144:23
De novo Predictions - 244:58
Reasoning in Knowledge Graphs45:43
Knowledge Graph46:01
Untitled47:41
Predictive Graph Queries48:28
Overview of Our Framework48:52
Example: Twitter - 150:15
Example: Twitter - 251:07
Example: Twitter - 351:08
Untitled52:33
Performance52:59
Conclusion54:14
Summary54:26
Conclusion - 155:18
Conclusion - 256:13
PhD Students57:24