Investigating Top-$k$ White-Box and Transferable Black-box Attack. (arXiv:2204.00089v1 [cs.LG])
Existing works have identified the limitation of top-$1$ attack success rate (ASR) as a metric to evaluate the attack strength...
Security World News
Category Added in a WPeMatico Campaign
Preventing Distillation-based Attacks on Neural Network IP. (arXiv:2204.00292v1 [cs.CR])
Neural networks (NNs) are already deployed in hardware today, becoming valuable intellectual property (IP) as many hours are invested in...
Investigating Top-$k$ White-Box and Transferable Black-box Attack. (arXiv:2204.00089v1 [cs.LG])
Existing works have identified the limitation of top-$1$ attack success rate (ASR) as a metric to evaluate the attack strength...
Preventing Distillation-based Attacks on Neural Network IP. (arXiv:2204.00292v1 [cs.CR])
Neural networks (NNs) are already deployed in hardware today, becoming valuable intellectual property (IP) as many hours are invested in...
Face identification by means of a neural net classifier. (arXiv:2204.00305v1 [cs.CV])
This paper describes a novel face identification method that combines the eigenfaces theory with the Neural Nets. We use the...
Face identification by means of a neural net classifier. (arXiv:2204.00305v1 [cs.CV])
This paper describes a novel face identification method that combines the eigenfaces theory with the Neural Nets. We use the...
Preventing Distillation-based Attacks on Neural Network IP. (arXiv:2204.00292v1 [cs.CR])
Neural networks (NNs) are already deployed in hardware today, becoming valuable intellectual property (IP) as many hours are invested in...
Face identification by means of a neural net classifier. (arXiv:2204.00305v1 [cs.CV])
This paper describes a novel face identification method that combines the eigenfaces theory with the Neural Nets. We use the...
A solution approach for the anonymous sharing of sensitive supply chain traceability data. (arXiv:2204.00317v1 [cs.CR])
Supply chain traceability systems have become central in many industries, however, traceability data is often commercially sensitive, and firms seek...
Robust and Accurate — Compositional Architectures for Randomized Smoothing. (arXiv:2204.00487v1 [cs.LG])
Randomized Smoothing (RS) is considered the state-of-the-art approach to obtain certifiably robust models for challenging tasks. However, current RS approaches...
