Results 31 to 40 of about 215,342 (268)
Target Training Does Adversarial Training Without Adversarial Samples
, 2021 arXiv admin note: text overlap with arXiv:2006 ...
openaire +2 more sources
Prior-Guided Adversarial Initialization for Fast Adversarial Training
, 2022 Fast adversarial training (FAT) effectively improves the efficiency of standard adversarial training (SAT). However, initial FAT encounters catastrophic overfitting, i.e.,the robust accuracy against adversarial attacks suddenly and dramatically decreases.
Jia, Xiaojun +6 more
openaire +2 more sources
Combining Adversaries with Anti-adversaries in Training
Proceedings of the AAAI Conference on Artificial Intelligence, 2023 Adversarial training is an effective learning technique to improve the robustness of deep neural networks. In this study, the influence of adversarial training on deep learning models in terms of fairness, robustness, and generalization is theoretically investigated under more general perturbation scope that different samples can have different ...
Zhou, Xiaoling, Yang, Nan, Wu, Ou
openaire +2 more sources
Universal Adversarial Training
Proceedings of the AAAI Conference on Artificial Intelligence, 2020 Standard adversarial attacks change the predicted class label of a selected image by adding specially tailored small perturbations to its pixels. In contrast, a universal perturbation is an update that can be added to any image in a broad class of images, while still changing the predicted class label.
Shafahi, Ali +5 more
openaire +3 more sources
EIFDAA: Evaluation of an IDS with function-discarding adversarial attacks in the IIoT
Heliyon, 2023 The complexity of the Industrial Internet of Things (IIoT) presents higher requirements for intrusion detection systems (IDSs). An adversarial attack is a threat to the security of machine learning-based IDSs.
Shiming Li +4 more
doaj +1 more source
Directional Adversarial Training for Robust Ownership-Based Recommendation System
IEEE Access, 2022 Machine learning algorithms are susceptible to cyberattacks, posing security problems in computer vision, speech recognition, and recommendation systems. So far, researchers have made great strides in adopting adversarial training as a defensive strategy.
Zhefu Wu +3 more
doaj +1 more source
Fast-M Adversarial Training Algorithm for Deep Neural Networks
Applied SciencesAlthough deep neural networks have been successfully applied in many fields, research studies show that neural network models are easily disrupted by small malicious inputs, greatly reducing their performance.
Yu Ma +4 more
doaj +1 more source
MAT: A Multi-strength Adversarial Training Method to Mitigate Adversarial Attacks
, 2018 Some recent works revealed that deep neural networks (DNNs) are vulnerable to so-called adversarial attacks where input examples are intentionally perturbed to fool DNNs.
Chen, Yiran +7 more
core +1 more source
Improving Adversarial Robustness via Distillation-Based Purification
Applied Sciences, 2023 Despite the impressive performance of deep neural networks on many different vision tasks, they have been known to be vulnerable to intentionally added noise to input images.
Inhwa Koo, Dong-Kyu Chae, Sang-Chul Lee
doaj +1 more source