Enhancing the Resilience of Graph Neural Networks to Topological Perturbations in Sparse Graphs.
Shuqi HeJun ZhuangDing WangLuyao PengJun SongPublished in: CoRR (2024)
Keyphrases
- neural network
- topological information
- directed acyclic
- graph theory
- graph representation
- graph structure
- graph theoretic
- labeled graphs
- graph matching
- graph construction
- gaussian graphical models
- directed graph
- weighted graph
- graph mining
- graph structures
- graph model
- random graphs
- adjacency matrix
- graph classification
- series parallel
- subgraph isomorphism
- graph theoretical
- undirected graph
- graph partitioning
- graph search
- pattern recognition
- bipartite graph
- densely connected
- graph databases
- graph properties
- graph data
- structural pattern recognition
- graph clustering
- maximum cardinality
- spanning tree
- dynamic graph
- reachability queries
- graphical models
- connected graphs
- graph representations
- maximum clique
- maximum common subgraph
- connected dominating set
- graph isomorphism
- planar graphs
- community discovery
- structured data
- connected components
- graph kernels
- maximal cliques
- attributed graphs
- random walk
- real world graphs
- graph patterns
- social graphs
- web graph
- minimum spanning tree
- high dimensional
- edge weights
- graph layout
- graph transformation
- adjacency graph
- proximity graph
- bounded degree
- dense subgraphs
- graph drawing
- inexact graph matching
- artificial neural networks
- small world
- polynomial time complexity
- reeb graph
- finding the shortest path
- topological map
- frequent subgraphs