Detecting Topology Attacks against Graph Neural Networks.
Senrong XuYuan YaoLiangyue LiWei YangFeng XuHanghang TongPublished in: CoRR (2022)
Keyphrases
- neural network
- fully connected
- artificial neural networks
- graph theory
- small world
- pattern recognition
- directed acyclic graph
- directed graph
- graph structure
- topological information
- countermeasures
- random walk
- detecting malicious
- graph representation
- graph model
- neural network model
- back propagation
- network intrusion detection systems
- recurrent neural networks
- structured data
- fuzzy logic
- undirected graph
- clustering algorithm
- activation function
- graph theoretic
- malicious attacks
- dos attacks
- security protocols
- topology preserving
- multi layer
- graph partitioning
- watermarking algorithm
- watermarking scheme
- weighted graph
- bipartite graph
- feed forward
- radial basis function
- image segmentation
- genetic algorithm