[HTML][HTML] Security in defect detection: A new one-pixel attack for fooling DNNs
Abstract The Industrial 5.0 Model integrates enabling technologies such as deep learning,
digital twins, and the meta-universe with new development concepts. However, model and …
digital twins, and the meta-universe with new development concepts. However, model and …
Exposing the Limits of Deepfake Detection using novel Facial mole attack: A Perceptual Black-Box Adversarial Attack Study
Recently, we have observed an exponential growth in highly realistic deepfake videos,
which are often used to spread disinformation, defame individuals, and even influence …
which are often used to spread disinformation, defame individuals, and even influence …
[PDF][PDF] Explanation-based Adversarial Detection with Noise Reduction
Abstract Deep Neural Networks (DNNs) have achieved tremendous success in various
tasks. However, DNNs expose uncertainty and unreliability against well-designed …
tasks. However, DNNs expose uncertainty and unreliability against well-designed …
Differential Evolution を用いたAdversarial Examples の生成における複数解探索
串田淳一 - 進化計算学会論文誌, 2023 - jstage.jst.go.jp
抄録 Over the past few years, deep neural networks (DNNs) have shown outstanding
performance in a wide range of domains. However, DNNs have been found to be vulnerable …
performance in a wide range of domains. However, DNNs have been found to be vulnerable …