A survey of safety and trustworthiness of deep neural networks: Verification, testing, adversarial attack and defence, and interpretability.
Xiaowei HuangDaniel KroeningWenjie RuanJames SharpYoucheng SunEmese ThamoMin WuXinping YiPublished in: Comput. Sci. Rev. (2020)
Keyphrases
- neural network
- safety analysis
- fault model
- fault injection
- java card
- formal verification
- test generation
- symbolic execution
- model checking
- artificial neural networks
- fault diagnosis
- fuzzy logic
- pattern recognition
- multi agent
- genetic algorithm
- trust relationships
- agent oriented
- neural nets
- recurrent neural networks
- feed forward
- multi layer
- neural network model
- back propagation
- social networks
- penetration testing
- training process
- software testing
- secret key
- test cases
- nuclear power plant
- prediction accuracy
- control system