Results 231 to 240 of about 217,331 (267)
Dimensionally constrained adversarial attack and defense in wind power forecasting. [PDF]
Min Y +5 more
europepmc +1 more source
Localized Query Attack Toward Transformer-Based Visible Object Detectors. [PDF]
Wang Y, Li A, Yang Z, Liu X.
europepmc +1 more source
Threats and vulnerabilities in artificial intelligence and agentic AI models. [PDF]
Radanliev P, Santos O, Maple C.
europepmc +1 more source
Some of the next articles are maybe not open access.
Related searches:
Related searches:
Boosting adversarial robustness via self-paced adversarial training
Neural Networks, 2023Adversarial training is considered one of the most effective methods to improve the adversarial robustness of deep neural networks. Despite the success, it still suffers from unsatisfactory performance and overfitting. Considering the intrinsic mechanism of adversarial training, recent studies adopt the idea of curriculum learning to alleviate ...
Lirong He +5 more
openaire +2 more sources
Adversarial Training With Anti-Adversaries
IEEE Transactions on Pattern Analysis and Machine IntelligenceAdversarial training is effective in improving the robustness of deep neural networks. However, existing studies still exhibit significant drawbacks in terms of the robustness, generalization, and fairness of models. In this study, we validate the importance of different perturbation directions (i.e., adversarial and anti-adversarial) and bounds from ...
Xiaoling Zhou, Ou Wu, Nan Yang
openaire +2 more sources
Adversarial training with Lookahead
2022Deep Learning wird in immer mehr sicherheitsrelevanten Bereichen wie zum Beispiel für autonomes Fahren oder in der automatischen Gesichtserkennung erfolgreich eingesetzt. Daher rückt der Sicherheitsaspekt von Deep Learning Algorithmen zunehmend in den Fokus der Forschung.
openaire +1 more source
Variational Adversarial Defense: A Bayes Perspective for Adversarial Training
IEEE Transactions on Pattern Analysis and Machine IntelligenceVarious methods have been proposed to defend against adversarial attacks. However, there is a lack of enough theoretical guarantee of the performance, thus leading to two problems: First, deficiency of necessary adversarial training samples might attenuate the normal gradient's back-propagation, which leads to overfitting and gradient masking ...
Chenglong Zhao +5 more
openaire +2 more sources

