2018 · 88 citations · 15 references
ComputingComputer ScienceCybersecurityApplication SecurityComputer SecurityMachine LearningDenial Of Service VulnerabilitiesSoftware Security EngineeringControl-flow HijackingDenial Of ServiceInformation Security RiskMachine-learning AttacksSecurity Risk ManagementVoice Recognition
Advances in deep learning algorithms overshadow their security risk in software implementations. This paper discloses a set of vulnerabilities in popular deep learning frameworks including Caffe, TensorFlow, and Torch. Contrary to the small code size of deep learning models, these deep learning frameworks are complex, and they heavily depend on numerous open source packages. This paper considers the risks caused by these vulnerabilities by studying their impact on common deep learning applications such as voice recognition and image classification. By exploiting these framework implementations, attackers can launch denial-of-service attacks that crash or hang a deep learning application, or control-flow hijacking attacks that lead to either system compromise or recognition evasions. The goal of this paper is to draw attention to software implementations and call for community collaborative effort to improve security of deep learning frameworks.
15
Yangqing Jia, Evan Shelhamer, Jeff Donahue et al. · 2014 · 11.2K citations
Explaining and Harnessing Adversarial Examples
Ian Goodfellow, Jonathon Shlens, Christian Szegedy · arXiv (Cornell University) · 2014 · 8.2K citations · Full text
Caffe: Convolutional Architecture for Fast Feature Embedding
Yangqing Jia, Evan Shelhamer, Jeff Donahue et al. · arXiv (Cornell University) · 2014 · 4.3K citations · Full text
The Limitations of Deep Learning in Adversarial Settings
Nicolas Papernot, Patrick McDaniel, Somesh Jha et al. · 2016 · 3.9K citations
Practical Black-Box Attacks against Machine Learning
Nicolas Papernot, Patrick McDaniel, Ian Goodfellow et al. · 2017 · 3.5K citations