Slide 1

Slide 1 text

1 (Supplement slides for reading paper) Semi-Supervised Graph Classification: A Hierarchical Graph Perspective(WWW19) izunan385

Slide 2

Slide 2 text

Li, Jia, et al. "Semi-Supervised Graph Classification: A Hierarchical Graph Perspective." (2019).

Slide 3

Slide 3 text

• Task Collect Class Prediction for unlabeled

Slide 4

Slide 4 text

• input each graph instance: g labeled graph set and unlabeled graph set graph instance adjacency matrix

Slide 5

Slide 5 text

• output IC(graph Instance Classifier) receives graph info and outputs instance representation matrix predicted class probability vector HC(Hierarchical Graph Classifier) receives all graph instance( ) representation from IC graph-graph adjacency matrix and outputs predicted class prob matrix for all

Slide 6

Slide 6 text

• Task Collect Class Prediction for unlabeled • Loss function labeled graph instances unlabeled graph instances

Slide 7

Slide 7 text

• Supervised Loss (for labeled graphs ) • Disagreement Loss(for unlabeled graphs ) Disagreement means IC and HC prediction mismatch.

Slide 8

Slide 8 text

No content

Slide 9

Slide 9 text

GCN W0: learnable parameter

Slide 10

Slide 10 text

GCN with self loop W0: learnable parameter

Slide 11

Slide 11 text

GCN(summarized) 0 https://www.experoinc.com/post/node-classification-by-graph-con network Adjacent/co-occurrence matrix has structure information. Propagation rule is learned during training.

Slide 12

Slide 12 text

https://docs.dgl.ai/tutorials/models/1_gnn/9_gat.html

Slide 13

Slide 13 text

Cautious Iteration

Slide 14

Slide 14 text

Cautious Iteration Here, sampling top confident prediction for each step

Slide 15

Slide 15 text

Active Iteration Disagreement means IC and HC prediction mismatch. Ask annotator for annotating class of graphs which HC and IC have top-disagreement with.