Results 91 to 100 of about 34,476 (259)

Batch-in-Batch: a new adversarial training framework for initial perturbation and sample selection

Complex & Intelligent Systems
Adversarial training methods commonly generate initial perturbations that are independent across epochs, and obtain subsequent adversarial training samples without selection.
Yinting Wu, Pai Peng, Bo Cai, Le Li
doaj   +1 more source

Three-Dimensional Reconstruction Pre-Training as a Prior to Improve Robustness to Adversarial Attacks and Spurious Correlation

Entropy
Ensuring robustness of image classifiers against adversarial attacks and spurious correlation has been challenging. One of the most effective methods for adversarial robustness is a type of data augmentation that uses adversarial examples during training.
Yutaro Yamada, Fred Weiying Zhang, Yuval Kluger, Ilker Yildirim   +3 more
doaj   +1 more source

PlantGFM: A Genomic Foundation Model for Discovery and Creation of Plant Genes

Advanced Science, EarlyView.
A plant genomic foundation model pre‐trained on 12 species enables both accurate gene prediction and de novo gene design. Through AI‐human knowledge screening, seven designed sequences showed transcriptional activity in plants, with two expressing stable proteins—demonstrating the first DNA‐RNA‐protein expression of LLM‐generated genes in plants and ...
Changhao Li, Qizhe Zhang, Hanchen Chen, Kepeng Lin, Chengfang Luo, Mengying Yang, Wei Xu, Fan Yao, Jianbing Yan, Qing‐Yong Yang, Xuehai Hu   +10 more
wiley   +1 more source

How Advanced Artificial Intelligence Technologies Shape Drug–Drug and Drug–Target Interaction Modeling

Advanced Science, EarlyView.
This review explores the convergence of artificial intelligence technologies in modeling drug–drug and drug–target interactions. By evaluating advanced feature engineering, architectural innovations, and learning paradigms reveals shared evolutionary trends and critical challenges, such as cold‐start settings and shortcut learning.
Xin Sun, Tong Wang
wiley   +1 more source

Clean, performance‐robust, and performance‐sensitive historical information based adversarial self‐distillation

IET Computer Vision
Adversarial training suffers from poor effectiveness due to the challenging optimisation of loss with hard labels. To address this issue, adversarial distillation has emerged as a potential solution, encouraging target models to mimic the output of the ...
Shuyi Li, Hongchao Hu, Shumin Huo, Hao Liang   +3 more
doaj   +1 more source

Semantics-Preserving Adversarial Training

CoRR, 2020
Preprint.
Wonseok Lee 0001, Hanbit Lee, Sang-goo Lee   +2 more
openaire   +2 more sources

AI‐Physics‐Experiment Trinity for Integrated Protein Dynamics Modeling

Advanced Science, EarlyView.
This review unites experiments, physics‐based simulations, and AI as a synergistic triad for protein dynamics modeling. It highlights integrative strategies, resolves sampling and forcefield bottlenecks, and outlines challenges and future directions for accurate, interpretable conformational ensemble prediction.
Chen Shi, Minying Low, Peng Xiu, Kresten Lindorff‐Larsen, Yong Wang   +4 more
wiley   +1 more source

Interpretable Adversarial Training for Text

CoRR, 2019
Generating high-quality and interpretable adversarial examples in the text domain is a much more daunting task than it is in the image domain. This is due partly to the discrete nature of text, partly to the problem of ensuring that the adversarial examples are still probable and interpretable, and partly to the problem of maintaining label invariance ...
Samuel Barham, Soheil Feizi
openaire   +2 more sources

Hardware‐Attentive Programmable Fourier Ptychography Enables Task‐Adaptive Label‐Free Virtual Staining

Advanced Science, EarlyView.
Task‐adaptive programmable optics enables label‐free virtual staining through optical‐attention‐guided acquisition and reconstruction. By optimizing wavelength, illumination angle, exposure time, and imaging depth, the framework learns task‐relevant optical measurements, generating clinically interpretable virtual stains with improved fidelity, non ...
Tianyue He, Wenyi Jing, Le Zhang, Qican Zhang, Tingdong Kou, Xin He, Yan Qiu, Yang Lu, Jian Cui, Yongfu Wen, Zhenrong Zheng, Hongying Zhang, Dayong Jin, Junfei Shen   +13 more
wiley   +1 more source

On Adversarial Robust Generalization of DNNs for Remote Sensing Image Classification

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Deep neural networks (DNNs)-based deep learning is an important technical support in the task of remote sensing image classification. But DNNs are susceptible to adversarial attacks.
Wei Xue, Yonghao Wang, Shaoquan Zhang, Xiao Zheng, Ping Zhong   +4 more
doaj   +1 more source