Detection of backdoor attacks using targeted universal adversarial perturbations for deep neural networks.
Yubin QuSong HuangXiang ChenXingya WangYongming YaoPublished in: J. Syst. Softw. (2024)
Keyphrases
- neural network
- automatic detection
- pattern recognition
- detection algorithm
- detection accuracy
- fuzzy logic
- object detection
- countermeasures
- attack detection
- malicious attacks
- detection method
- back propagation
- self organizing maps
- network attacks
- false alarms
- multi layer
- feed forward
- detection rate
- event detection
- false positives
- data mining
- anomaly detection
- genetic algorithm