Defensive dropout for hardening deep neural networks under adversarial attacks.
Siyue WangXiao WangPu ZhaoWujie WenDavid R. KaeliSang Peter ChinXue LinPublished in: ICCAD (2018)
Keyphrases
- neural network
- pattern recognition
- countermeasures
- multi agent
- fuzzy logic
- neural network model
- artificial neural networks
- fuzzy systems
- malicious users
- traffic analysis
- security threats
- image watermarking
- computer security
- self organizing maps
- genetic algorithm
- multi layer
- training process
- neural nets
- security mechanisms
- recurrent neural networks
- deep learning
- dos attacks
- feed forward
- back propagation
- denial of service attacks
- malicious attacks