Results 121 to 130 of about 27,717 (260)
Boosting adversarial robustness via feature refinement, suppression, and alignment
Complex & Intelligent SystemsDeep neural networks are vulnerable to adversarial attacks, bringing high risk to numerous security-critical applications. Existing adversarial defense algorithms primarily concentrate on optimizing adversarial training strategies to improve the ...
Yulun Wu +6 more
doaj +1 more source
Adversarial Examples Are Not Bugs, They Are Superposition
CoRRAdversarial examples -- inputs with imperceptible perturbations that fool neural networks -- remain one of deep learning's most perplexing phenomena despite nearly a decade of research. While numerous defenses and explanations have been proposed, there is no consensus on the fundamental mechanism.
Liv Gorton, Owen Lewis
openaire +2 more sources
AI &Innovation, EarlyView.ABSTRACT The rapid evolution of the Internet of Things (IoT) has significantly advanced the field of electrocardiogram (ECG) monitoring, enabling real‐time, remote, and patient‐centric cardiac care. This paper presents a comprehensive survey of AI assisted IoT‐based ECG monitoring systems, focusing on the integration of emerging technologies such as ...
Amrita Choudhury +2 more
wiley +1 more source
Dual-Mode Method for Generating Adversarial Examples to Attack Deep Neural Networks
IEEE AccessDeep neural networks yield desirable performance in text, image, and speech classification. However, these networks are vulnerable to adversarial examples. An adversarial example is a sample generated by inserting a small amount of noise into an original
Hyun Kwon, Sunghwan Kim
doaj +1 more source
Playing in the Dark: Invisible Chess as a Laboratory for Strategic AI
AI &Innovation, EarlyView.This paper shows that strategic AI evaluated on perfect‐information benchmarks can be brittle in real adversarial settings. By using invisible chess as a benchmark for hidden state and deception, it argues for stricter testing, human oversight, and more cautious governance of high‐stakes AI systems.
Paolo Ciancarini
wiley +1 more source
Approaching Adversarial Example Classification with Chaos Theory. [PDF]
Entropy (Basel), 2020 Pedraza A, Deniz O, Bueno G.
europepmc +1 more source
Australian Journal of Social Issues, EarlyView.ABSTRACT Improving access to legal services for Indigenous, migrant and refugee women is critical to addressing family violence. In this context, Family Dispute Resolution (FDR) has long been discussed as a solution for separating families. This paper presents key findings of a research evaluation of an Australian Government $8.37 million pilot project
Siobhan McDonnell, Alyson Wright
wiley +1 more source
Friend-Guard Textfooler Attack on Text Classification System
IEEE AccessDeep neural networks provide good performance for image classification, text classification, speech classification, and pattern analysis. However, such neural networks are vulnerable to adversarial examples.
Hyun Kwon
doaj +1 more source
Applications in Plant Sciences, EarlyView.Abstract Premise Accurate species identification is crucial for ecological restoration and can be especially challenging for understudied non‐model species. Quercus garryana is the only native oak species in the Pacific Northwest and is an important component of the endangered oak savanna ecosystem.
Ziqi Xie, Tan Bahar, Keith Karoly
wiley +1 more source
Journal of the Association for Information Science and Technology, EarlyView.Abstract The integrity of scholarly communication depends critically on the accuracy and verifiability of cited references. Citations enable readers to trace prior work, assess evidence, and situate new contributions within the existing literature. However, concerns have emerged regarding the presence of references in published papers that cannot be ...
Chengcheng Han +3 more
wiley +1 more source