Skip to content

Simplicial Weisfeiler and Lehman for graph isomorphism test

Notifications You must be signed in to change notification settings

2019mohamed/SWL

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 

Repository files navigation

SWL

Simplicial Weisfeiler and Lehman for graph isomorphism test from Weisfeiler and Lehman Go Topological: Message Passing Simplicial Networks

perfect graphs with simpilical complex(i.e., cliques) GitHub Logo GitHub Logo GitHub Logo

(Very simple survay) In Message Passing Simplicial Networks(i.e., hight dim.) or Message Passing framwork, we have to define aggregation operator over neighbors of vertex. So, here, aggregation over faces and cofaces. we can aggrate over a lot of neighbors sets to get expressive poweful operator. However, SWL > 3-WL, rGIN model(i.e., GIN with random vertex features) >= 1-WL. Also, there exists "Graph Neural Network Expressivity via Subgraph Isomorphism Counting" model(i.e., GIN with structual vertex or edge features) > 3-WL, and "CRaWl (Convolutional Neural Networks for Random Walks)" model >= k-WL >= k-GNN !!! The story is not over...

About

Simplicial Weisfeiler and Lehman for graph isomorphism test

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages