Results 141 to 150 of about 243,531 (318)

Boosting adversarial robustness via feature refinement, suppression, and alignment

Complex & Intelligent Systems
Deep 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, Yanming Guo, Dongmei Chen, Tianyuan Yu, Huaxin Xiao, Yuanhao Guo, Liang Bai   +6 more
doaj   +1 more source

Dual-Mode Method for Generating Adversarial Examples to Attack Deep Neural Networks

IEEE Access
Deep 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

Real‐Time Multicolor Fluorescence Microscopy via Cross‐Channel Acquisition and Deep‐Learning‐Based Inference

Advanced Intelligent Discovery, EarlyView.
Sequential multicolor fluorescence imaging in dynamic microsystems is constrained by acquisition speed and excitation dose. This study introduces a real‐time framework to reconstruct spectrally separated channels from reduced cross‐channel acquisitions (frames containing mixed spectral contributions).
Juan J. Huaroto, Yuxin Jin, Nicholas Roy, Sarthak Misra   +3 more
wiley   +1 more source

Friend-Guard Textfooler Attack on Text Classification System

IEEE Access
Deep 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

Robust Text CAPTCHAs Using Adversarial Examples [PDF]

, 2021
Rulin Shao, Zhouxing Shi, Jinfeng Yi, Pin‐Yu Chen, Cho‐Jui Hsieh   +4 more
openalex   +1 more source

Cell Segmentation Beyond 2D—A Review of the State‐of‐the‐Art

Advanced Intelligent Discovery, EarlyView.
Cell segmentation underpins many biological image analysis tasks, yet most deep learning methods remain limited to 2D despite the inherently 3D nature of cellular processes. This review surveys segmentation approaches beyond 2D, comparing 2.5D and fully 3D methods, analyzing 31 models and 32 volumetric datasets, and introducing a unified reference ...
Fabian Schmeisser, Haseeb Muhammad, Junaid Younas, Gillian Lovell, Bianca Migliori, Andreas Dengel, Sheraz Ahmed   +6 more
wiley   +1 more source

RAMS: Residual‐Based Adversarial‐Gradient Moving Sample Method for Scientific Machine Learning in Solving Partial Differential Equations

Advanced Intelligent Discovery, EarlyView.
We propose a residual‐based adversarial‐gradient moving sample (RAMS) method for scientific machine learning that treats samples as trainable variables and updates them to maximize the physics residual, thereby effectively concentrating samples in inadequately learned regions.
Weihang Ouyang, Min Zhu, Wei Xiong, Si‐Wei Liu, Lu Lu   +4 more
wiley   +1 more source

Composition‐Aware Cross‐Sectional Integration for Spatial Transcriptomics

Advanced Intelligent Discovery, EarlyView.
Multi‐section spatial transcriptomics demands coherent cell‐type deconvolution, domain detection, and batch correction, yet existing pipelines treat these tasks separately. FUSION unifies them within a composition‐aware latent framework, modeling reads as cell‐type–specific topics and clustering in embedding space.
Qishi Dong, Zijian Huang, Xuanwu Wang, Zhenghui Feng, Wei Liu, Bingding Huang   +5 more
wiley   +1 more source

Generating Image Adversarial Examples by Embedding Digital Watermarks [PDF]

, 2020
Yuexin Xiang, Tiantian Li, Wei Ren, Tianqing Zhu, Kim‐Kwang Raymond Choo   +4 more
openalex   +1 more source

BeamAttack: Generating High-quality Textual Adversarial Examples through Beam Search and Mixed Semantic Spaces [PDF]

, 2023
Hai Zhu, Zhao, Qingyang, Yuren Wu
openalex   +1 more source