Results 181 to 190 of about 26,718 (222)
Energy-Efficient and Adversarially Resilient Underwater Object Detection via Adaptive Vision Transformers. [PDF]
Sensors (Basel)Li L, Zhang G, Zhou Y.
europepmc +1 more source
Attenuation correction of cardiac <sup>82</sup>Rb pet using deep learning generated synthetic CT. [PDF]
EJNMMI PhysJørgensen K +4 more
europepmc +1 more source
SecMLOps: A comprehensive framework for integrating security throughout the machine learning operations lifecycle. [PDF]
Empir Softw EngZhang X, Zhao P, Jaskolka J, Li H, Lu R.
europepmc +1 more source
Traversing the subspace of adversarial patches
Machine Vision and ApplicationsAbstract Despite ongoing research on the topic of adversarial examples in deep learning for computer vision, some fundamentals of the nature of these attacks remain unclear. As the manifold hypothesis posits, high-dimensional data tends to be part of a low-dimensional manifold.
Jens Bayer +2 more
exaly +4 more sources
Some of the next articles are maybe not open access.
Related searches:
Related searches:
A data independent approach to generate adversarial patches
Machine Vision and Applications, 2021Deep neural networks are vulnerable to adversarial examples, i.e., carefully perturbed inputs designed to mislead the network at inference time. Recently, adversarial patch, with perturbations confined to a small and localized patch, emerged for its easy accessibility in real-world attack.
Xingyu Zhou 0002 +4 more
openaire +1 more source
Enhancing the Transferability of Adversarial Examples with Random Patch
Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence, 2022Adversarial examples can fool deep learning models, and their transferability is critical for attacking black-box models in real-world scenarios. Existing state-of-the-art transferable adversarial attacks tend to exploit intrinsic features of objects to generate adversarial examples.
Yaoyuan Zhang +5 more
openaire +1 more source
Jedi: Entropy-Based Localization and Removal of Adversarial Patches
2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023 Real-world adversarial physical patches were shown to be successful in compromising state-of-the-art models in a variety of computer vision applications. Existing defenses that are based on either input gradient or features analysis have been compromised by recent GAN-based attacks that generate naturalistic patches.
Anouar Ben Khalifa +2 more
exaly +5 more sources
An information-theoretic perspective of physical adversarial patches [PDF]
Neural NetworksReal-world adversarial patches were shown to be successful in compromising state-of-the-art models in various computer vision applications. Most existing defenses rely on analyzing input or feature level gradients to detect the patch. However, these methods have been compromised by recent GAN-based attacks that generate naturalistic patches.
Ihsen Alouani +2 more
exaly +5 more sources