One Pixel Attack for Fooling Deep Neural Networks
Recent research has revealed that the output of Deep Neural Networks (DNN) can be easily altered by adding relatively small perturbations to the input vector. In this paper, we ...
Publications
1 shownFrequent Co-Authors
Researcher Info
- h-index
- 1
- Publications
- 2
- Citations
- 3,156
- Institution
- Advanced Telecommunications Research Institute International
External Links
Identifiers
- ORCID
- 0000-0003-4621-1674
Impact Metrics
h-index
1
h-index: Number of publications with at least h citations each.