Sparse Vicious Attacks on Graph Neural Networks.
Giovanni TrappoliniValentino MaiorcaSilvio SeverinoEmanuele RodolàFabrizio SilvestriGabriele TolomeiPublished in: CoRR (2022)
Keyphrases
- neural network
- graph theory
- gaussian graphical models
- malicious attacks
- fuzzy logic
- graph representation
- random walk
- graph structure
- directed graph
- genetic algorithm
- pattern recognition
- countermeasures
- multi layer
- graph matching
- high dimensional
- artificial neural networks
- back propagation
- security threats
- sparse data
- directed acyclic graph
- weighted graph
- connected components
- neural nets
- fault diagnosis
- graph mining
- graph databases
- feed forward
- watermarking algorithm
- digital images
- compressive sensing
- security protocols
- image segmentation
- security risks
- neural network model