Bounding the Expected Robustness of Graph Neural Networks Subject to Node Feature Attacks.
Yassine AbbahaddouSofiane EnnadirJohannes F. LutzeyerMichalis VazirgiannisHenrik BoströmPublished in: ICLR (2024)
Keyphrases
- neural network
- graph structure
- directed graph
- pattern recognition
- finding the shortest path
- digital image watermarking
- tree structure
- image watermarking scheme
- random walk
- graph representation
- countermeasures
- weighted graph
- directed acyclic graph
- overlapping communities
- edge weights
- graph model
- neural network model
- image features
- connected components
- path length
- vector representation
- back propagation
- upper bound
- nodes of a graph
- tree traversal
- graph theory
- fuzzy logic
- semi fragile watermarking