Results 51 to 60 of about 219,753 (266)
Multi-Class Triplet Loss With Gaussian Noise for Adversarial Robustness
IEEE Access, 2020 Deep Neural Networks (DNNs) classifiers performance degrades under adversarial attacks, such attacks are indistinguishably perturbed relative to the original data.
Benjamin Appiah +4 more
doaj +1 more source
A Robust Method to Protect Text Classification Models against Adversarial Attacks
Proceedings of the International Florida Artificial Intelligence Research Society Conference, 2022 Text classification is one of the main tasks in natural language processing. Recently, adversarial attacks have shown a substantial negative impact on neural network-based text classification models. There are few defenses to strengthen model predictions
BALA MALLIKARJUNARAO GARLAPATI +2 more
doaj +1 more source
All‐Optical Reconfigurable Physical Unclonable Function for Sustainable Security
Advanced Materials, EarlyView.An all‐optical reconfigurable physical unclonable function (PUF) is demonstrated using plasmonic coupling–induced sintering of optically trapped gold nanoparticles, where Brownian motion serves as a robust entropy source. The resulting optical PUF exhibits high encoding density, strong resistance to modeling attacks, and practical authentication ...
Jang‐Kyun Kwak +4 more
wiley +1 more source
Adversarial Dropout for Supervised and Semi-supervised Learning
, 2017 Recently, the training with adversarial examples, which are generated by adding a small but worst-case perturbation on input examples, has been proved to improve generalization performance of neural networks. In contrast to the individually biased inputs
Moon, Il-Chul +3 more
core +1 more source
Splitting the Difference on Adversarial Training
CoRR, 2023 The existence of adversarial examples points to a basic weakness of deep neural networks. One of the most effective defenses against such examples, adversarial training, entails training models with some degree of robustness, usually at the expense of a degraded natural accuracy.
Matan Levi, Aryeh Kontorovich
openaire +3 more sources
Advanced Materials Technologies, EarlyView.The article overviews past and current efforts on caloric materials and systems, highlighting the contributions of Ames National Laboratory to the field. Solid‐state caloric heat pumping is an innovative method that can be implemented in a wide range of cooling and heating applications.
Agata Czernuszewicz +5 more
wiley +1 more source
On the existence of solutions to adversarial training in multiclass classification
European Journal of Applied MathematicsAdversarial training is a min-max optimization problem that is designed to construct robust classifiers against adversarial perturbations of data. We study three models of adversarial training in the multiclass agnostic-classifier setting.
Nicolás García Trillos +2 more
doaj +1 more source
Squeeze Training for Adversarial Robustness
, 2022 The vulnerability of deep neural networks (DNNs) to adversarial examples has attracted great attention in the machine learning community. The problem is related to non-flatness and non-smoothness of normally obtained loss landscapes. Training augmented with adversarial examples (a.k.a., adversarial training) is considered as an effective remedy.
Qizhang Li +3 more
openaire +3 more sources
Group-Structured Adversarial Training
CoRR, 2021 Robust training methods against perturbations to the input data have received great attention in the machine learning literature. A standard approach in this direction is adversarial training which learns a model using adversarially-perturbed training samples.
Farzan Farnia +3 more
openaire +2 more sources
Learning Highly Dynamic Skills Transition for Quadruped Jumping Through Constrained Space
Advanced Robotics Research, EarlyView.A quadruped robot masters dynamic jumps through constrained spaces with animal‐inspired moves and intelligent vision control. This hierarchical learning approach combines imitation of biological agility with real‐time trajectory planning. Although legged animals are capable of performing explosive motions while traversing confined spaces, replicating ...
Zeren Luo +6 more
wiley +1 more source