Adversarial Attacks on Neural Network with Batch Dimensions Perturbation and Manhattan-Distance Constraints
نویسندگان
چکیده
منابع مشابه
Adversarial Attacks on Neural Network Policies
Machine learning classifiers are known to be vulnerable to inputs maliciously constructed by adversaries to force misclassification. Such adversarial examples have been extensively studied in the context of computer vision applications. In this work, we show adversarial attacks are also effective when targeting neural network policies in reinforcement learning. Specifically, we show existing ad...
متن کاملDetecting Adversarial Attacks on Neural Network Policies with Visual Foresight
Deep reinforcement learning has shown promising results in learning control policies for complex sequential decision-making tasks. However, these neural network-based policies are known to be vulnerable to adversarial examples. This vulnerability poses a potentially serious threat to safety-critical systems such as autonomous vehicles. In this paper, we propose a defense mechanism to defend rei...
متن کاملAdversarial Generative Nets: Neural Network Attacks on State-of-the-Art Face Recognition
In this paper we show that misclassification attacks against face-recognition systems based on deep neural networks (DNNs) are more dangerous than previously demonstrated, even in contexts where the adversary can manipulate only her physical appearance (versus directly manipulating the image input to the DNN). Specifically, we show how to create eyeglasses that, when worn, can succeed in target...
متن کاملModeling Distributed Network Attacks with Constraints
In this work we demonstrate how to model and perform the detection of Distributed Network attacks using NeMODe, a declarative system for Computer Network Intrusion Detection which provides a declarative Domain Specific Language for describing computer network intrusion signatures which span several network packets by stating constraints over network packets, thus, describing relations between s...
متن کاملModelling distributed network attacks with constraints
NeMODe is a declarative system for computer network intrusion detection, providing a declarative domain specific language for describing network intrusion signatures which can span several network packets, by stating constraints over network packets, describing relations between several packets in a declarative and expressive way. It provides several back-end detection mechanisms, all based on ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: International Journal of Machine Learning and Computing
سال: 2022
ISSN: ['2010-3700']
DOI: https://doi.org/10.18178/ijmlc.2022.12.1.1073