Subgraph-Based Networks for Expressive, Efficient, and Domain-Independent Graph Learning
IEEE Signal Processing Society via YouTube
Overview
Syllabus
Introduction
Learning on graphs
Explicit power
Color refinement
Limitless exclusivity
Why exclusivity matters
Goal
Recipe
What is a suitable neural network
Equivariance
Equivariant
Benefits
Two kinds of symmetry
DSS
Architecture
Large graphs
Theoretical analysis
Experimental analysis
Evaliant graph networks
Nodebased policies
Tensors
Intuition
Other approaches
Sun architecture
Brain
Experiment
Summary
Conclusion
Questions
Taught by
IEEE Signal Processing Society