Towards Robust DNNs: An Taylor Expansion-Based Method for Generating Powerful Adversarial Examples.
Yaguan QianXi-Ming ZhangBin WangWei LiJian-Hai ChenWujie ZhouJing-Sheng LeiPublished in: CoRR (2020)
Keyphrases
- high precision
- preprocessing
- detection method
- highly accurate
- significant improvement
- dynamic programming
- computationally efficient
- computational complexity
- classification method
- segmentation method
- mutual information
- neural network
- synthetic data
- clustering method
- input data
- highly efficient
- detection algorithm
- classification accuracy
- computational cost
- experimental evaluation
- image retrieval
- genetic algorithm