Reverse engineering imperceptible backdoor attacks on deep neural networks for detection and training set cleansing.
Zhen XiangDavid J. MillerGeorge KesidisPublished in: Comput. Secur. (2021)
Keyphrases
- reverse engineering
- neural network
- training set
- software engineering
- watermarking scheme
- software maintenance
- dynamic analysis
- malicious code
- gene regulatory networks
- data sets
- attack detection
- object oriented
- test set
- conceptual schema
- software evolution
- reverse engineer
- computer aided design
- watermarking technique
- watermarking algorithm
- software engineers
- database
- watermark detection
- program understanding
- training data
- digital watermarking
- artificial neural networks
- anomaly detection
- supervised learning
- data hiding
- platform independent
- high dimensional
- xml documents
- databases
- fault diagnosis