检索结果-内蒙古大学图书馆

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Liu, Xuan Cai, Siqi Zhou, Qihua Guo, Song Li, Ruibin Lin, Kaiwei The Hong Kong Polytechnic University Hong Kong Hubei Key Laboratory of Transportation Internet of Things Wuhan University of Technology Wuhan China College of Computer Science and Software Engineering Shenzhen University Shenzhen China Department of Computer Science and Engineering The Hong Kong University of Science and Technology Hong Kong

ISBN: (纸本)157735897X

Perturbation-based mechanisms, such as differential privacy, mitigate gradient leakage attacks by introducing noise into the gradients, thereby preventing attackers from reconstructing clients’ private data from the leaked gradients. However, can gradient perturbation protection mechanisms truly defend against all gradient leakage attacks? In this paper, we present the first attempt to break the shield of gradient perturbation protection in Federated Learning for the extraction of private information. We focus on common noise distributions, specifically Gaussian and Laplace, and apply our approach to DNN and CNN models. We introduce Mjölnir, a perturbation-resilient gradient leakage attack that is capable of removing perturbations from gradients without requiring additional access to the original model structure or external data. Specifically, we leverage the inherent diffusion properties of gradient perturbation protection to develop a novel diffusion-based gradient denoising model for Mjölnir. By constructing a surrogate client model that captures the structure of perturbed gradients, we obtain crucial gradient data for training the diffusion model. We further utilize the insight that monitoring disturbance levels during the reverse diffusion process can enhance gradient denoising capabilities, allowing Mjölnir to generate gradients that closely approximate the original, unperturbed versions through adaptive sampling steps. Extensive experiments demonstrate that Mjölnir effectively recovers the protected gradients and exposes the Federated Learning process to the threat of gradient leakage, achieving superior performance in gradient denoising and private data recovery. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Multipath matching pursuit using a cross-validation technique for sparse direction-of-arrival estimation with an acoustic vector array

引用

Chinese Journal of Acoustics 2023年第2期42卷 153-170页

作者： TAN Peng HU Bo ZHANG Youwen WU Yuquan YANG Desen Acoustic Science and Technology Laboratory Harbin Engineering UniversityHarbin 150001 Key Laboratory of Marine Information Acquisition and Security(Harbin Engineering University) Ministry of Industry and Information TechnologyHarbin 150001 College of Underwater Acoustic Engineering Harbin Engineering UniversityHarbin 150001 School of Electronics and Information Jiangsu University of Science and TechnologyZhenjiang 212100 Science and Technology on Integrated Information System Laboratory Institute of SoftwareChinese Academy of SciencesBeijing 100190

High-resolution direction-of-arrival(DOA)estimations and the starboard ambiguity of moving underwater targets have always been key issues in underwater acoustic array signal *** with sound pressure arrays,vector arrays have natural advantages with respect to solving the starboard ambiguity problem and obtaining higher processing *** high-resolution DOA estimation methods such as Capon have disadvantages such as being unable to resolve coherent sources,requiring multiple snapshot processing,and being sensitive to array manifold ***-resolution DOA estimation and the starboard ambiguity of moving underwater targets have always been challenging research *** one hand,maneuvering underwater targets reduce the coherence time of the received signals,which ultimately leads to poor performance when using high-resolution DOA estimation technologies based on the covariance matrix of the received *** the other hand,traditional DOA estimation technologies based on sound pressure arrays have the problem of port and starboard ambiguity,which can be solved by maneuvering the sonar ***,maneuvering the sonar platform can impair the coherence of the received signal,on which some algorithms *** approach greatly limits the combat effectiveness and performance of the *** the aforementioned problems and taking advantage of the target sparsity,a cross-validation multipath matching pursuit technique based on the sparse DOA estimation of an acoustic vector array is proposed in this article for sonar *** proposed algorithm uses cross-validation technology to achieve a sparse DOA estimation with an unknown number of targets in a sonar observation *** with the conventional acoustic vector array-based Capon algorithm,the proposed algorithm can achieve a sparse DOA estimation and high-resolution capability with small numbers of snapshots or even single *** effectiveness of the proposed

关键词： multipath estimation underwater

来源：评论

学校读者我要写书评

暂无评论

Breaking Barriers in Physical-World Adversarial Examples: Improving Robustness and Transferability via Robust Feature 39

Breaking Barriers in Physical-World Adversarial Examples: Im...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Wang, Yichen Chou, Yuxuan Zhou, Ziqi Zhang, Hangtao Wan, Wei Hu, Shengshan Li, Minghui National Engineering Research Center for Big Data Technology and System China Services Computing Technology and System Lab China Cluster and Grid Computing Lab China Hubei Engineering Research Center on Big Data Security China Hubei Key Laboratory of Distributed System Security China School of Cyber Science and Engineering Huazhong University of Science and Technology China School of Computer Science and Technology Huazhong University of Science and Technology China School of Software Engineering Huazhong University of Science and Technology China

ISBN: (纸本)157735897X

As deep neural networks (DNNs) are widely applied in the physical world, many researches are focusing on physical-world adversarial examples (PAEs), which introduce perturbations to inputs and cause the model's incorrect outputs. However, existing PAEs face two challenges: unsatisfactory attack performance (i.e., poor transferability and insufficient robustness to environment conditions), and difficulty in balancing attack effectiveness with stealthiness, where better attack effectiveness often makes PAEs more perceptible. In this paper, we explore a novel perturbation-based method to overcome the challenges. For the first challenge, we introduce a strategy Deceptive RF injection based on robust features (RFs) that are predictive, robust to perturbations, and consistent across different models. Specifically, it improves the transferability and robustness of PAEs by covering RFs of other classes onto the predictive features in clean images. For the second challenge, we introduce another strategy Adversarial Semantic Pattern Minimization, which removes most perturbations and retains only essential adversarial patterns in AEs. Based on the two strategies, we design our method Robust Feature Coverage Attack (RFCoA), comprising Robust Feature Disentanglement and Adversarial Feature Fusion. In the first stage, we extract target class RFs in feature space. In the second stage, we use attention-based feature fusion to overlay these RFs onto predictive features of clean images and remove unnecessary perturbations. Experiments show our method's superior transferability, robustness, and stealthiness compared to existing state-of-the-art methods. Additionally, our method's effectiveness can extend to Large Vision-Language Models (LVLMs), indicating its potential applicability to more complex tasks. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Convolutional Neural Network based Digital Twin of Rolling Bearings for CNC Machine Tools in Cloud Computing 10

Convolutional Neural Network based Digital Twin of Rolling B...

引用

10th International Conference on Dependable Systems and Their Applications, DSA 2023

作者： Wang, Zhaoming Gong, Siqian Liu, Yang Beijing Jiaotong University School of Mechanical Electronic and Control Engineering Beijing China Nanjing University State Key Laboratory for Novel Software Technology Nanjing China

ISBN: (纸本)9798350304770

CNC machine tool is the mother of industry, and the realization of Intelligent CNC machine tool is an important link to promote the development of intelligent manufacturing. The fault diagnosis of key components of CNC machine tools is very important for the machining process. The traditional fault diagnosis methods have problems such as poor real-time performance and inaccurate prediction. Therefore, with the help of digital twin technology and cloud computing, this paper proposes a five-dimensional digital twin system architecture of CNC machine tool rolling bearing based on cloud computing to solve the above problems. And a deep learning technology for data processing is established, which is a hybrid network structure based on CNN-LSTM. Taking the fault diagnosis and prediction of rolling bearing as an example, it is verified that the prediction effect of the algorithm is good, and it can be used for system data processing. © 2023 IEEE.

关键词： Cloud computing

来源：评论

学校读者我要写书评

暂无评论

CDGNet: A Cross-Time Dynamic Graph-Based Deep Learning Model for Vehicle-Based Traffic Speed Forecasting

IEEE Transactions on Intelligent Vehicles

引用

IEEE Transactions on Intelligent Vehicles 2024年 1-14页

作者： Fang, Yuchen Luo, Haiyong Zhao, Fang Sun, Poly Z.H. Qin, Yanjun Zeng, Liang Hui, Bo Wang, Chenxing School of Computer Science National Pilot Software Engineering School Beijing University of Posts and Telecommunications Beijing China Beijing Key Laboratory of Mobile Computing and Pervasive Device Institute of Computing Technology Chinese Academy of Sciences Beijing China Department of Industrial Engineering School of Mechanical Engineering Shanghai Jiao Tong University Shanghai China Tsinghua University Beijing China University of Tulsa Tulsa USA

Vehicle-based traffic speed forecasting aims to predict the average speed of vehicles on the road in the future, which is an essential side information in intelligent vehicles and beneficial to safe autonomous driving, yet is very challenging due to the complex and dynamic spatio-temporal dependencies in real-world vehicle-based traffic. To extract intricate correlations among multiple vehicle-based speed time series, previous methods have used graph convolution networks. However, the conventional static and dynamic graphs fail to reflect the traffic evolution and the hysteresis spatial influence caused by vehicle movement. To address this issue, we propose a novel cross-time dynamic graph-based deep learning model, named CDGNet, for vehicle-based traffic speed forecasting. The model is able to effectively capture hysteresis spatial dependencies between each time slice and its historical time slices through the cross-time dynamic graph-based GCN. Meanwhile, a gating mechanism is integrated into our cross-time dynamic graph, which conforms to the sparse correlation in the real world. Besides, GCNs are incorporated into a novel encoder-decoder architecture to forecast multi-step speed. Experimental results on three real-world vehicle-based traffic speed datasets demonstrate the superiority of our CDGNet over various state-of-the-art spatio-temporal forecasting methods and the effectiveness of each component. We additionally provide a visualization of our cross-time dynamic graph to show the capability of assisting intelligent vehicles to avoid congestion. IEEE

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

CA-Edit: Causality-Aware Condition Adapter for High-Fidelity Local Facial Attribute Editing 39

CA-Edit: Causality-Aware Condition Adapter for High-Fidelity...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Xian, Xiaole He, Xilin Niu, Zenghao Zhang, Junliang Xie, Weicheng Song, Siyang Yu, Zitong Shen, Linlin Computer Vision Institute School of Computer Science & Software Engineering Shenzhen University China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University China Guangdong Provincial Key Laboratory of Intelligent Information Processing China University of Exeter United Kingdom Great Bay University China

ISBN: (纸本)157735897X

For efficient and high-fidelity local facial attribute editing, most existing editing methods either require additional finetuning for different editing effects or tend to affect beyond the editing regions. Alternatively, inpainting methods can edit the target image region while preserving external areas. However, current inpainting methods still suffer from the generation misalignment with facial attributes description and the loss of facial skin details. To address these challenges, (i) a novel data utilization strategy is introduced to construct datasets consisting of attribute-text-image triples from a data-driven perspective, (ii) a Causality-Aware Condition Adapter is proposed to enhance the contextual causality modeling of specific details, which encodes the skin details from the original image while preventing conflicts between these cues and textual conditions. In addition, a Skin Transition Frequency Guidance technique is introduced for the local modeling of contextual causality via sampling guidance driven by low-frequency alignment. Extensive quantitative and qualitative experiments demonstrate the effectiveness of our method in boosting both fidelity and editability for localized attribute editing. Our codes will be made publicly available. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Adaptive Occlusion Face Recognition Feature Fusion Network Based on Precise Guidance Mask Extraction

引用

Journal of Shanghai Jiaotong University (Science) 2025年 1-12页

作者： Ye, Jihua Zhang, Lei Wang, Chao Tu, Qixuan Zheng, Rehua Huang, Huiyuan Jiang, Aiwen School of Computer and Information Engineering Jiangxi Normal University Nanchang330022 China Jiangxi Provincial Key Laboratory for High Performance Computing Jiangxi Normal University Nanchang330022 China State International Science & Technology Cooperation Base of Networked Supporting Software Jiangxi Normal University Nanchang330022 China

The uncertainty in the position and size of occluding objects greatly affects the extraction of identity features in facial recognition, which is a challenge that existing methods fail to effectively address. To tackle this issue, this paper introduces a novel feature fusion network for occluded facial recognition that adaptively handles arbitrary occlusions and accurately extracts occlusion information masks. The network effectively mitigates the impact of occlusions on facial identity features through a feature fusion module. The proposed method employs an occlusion classification module and an occlusion classification loss to precisely guide the extraction of accurate occlusion masks. Subsequently, an occlusion-aware hybrid attention module is introduced, which focuses on the relationship between facial features and occlusion masks across channel and spatial dimensions. This fusion strategy enables the extraction of facial identity features that are resilient to occlusions. Experimental results indicate that the proposed method holds a significant advantage in handling facial recognition tasks with arbitrary occlusions. On the LFW and MF1 synthetic occlusion datasets, the average accuracy rates reached 93.19% and 72.57%, respectively, showing improvements of 1.7 and 2.39 percent points over mainstream methods. © Shanghai Jiao Tong University 2025.

关键词： A feature fusion occluded facial recognition occlusion mask occlusion perception TP391.41

来源：评论

学校读者我要写书评

暂无评论

Transformer-based multi-level attention integration network for video saliency prediction

引用

Multimedia Tools and Applications 2025年第13期84卷 11833-11854页

作者： Tan, Rui Sun, Minghui Liang, Yanhua Software College Jilin University Qianjin Street Jilin Changchun130012 China College of Computer Science and Technology Jilin University Qianjin Street Jilin Changchun130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Qianjin Street Jilin Changchun130012 China

Most existing models for video saliency prediction heavily rely on 3D convolutional operations to extract spatio-temporal features. However, it is worth noting that 3D convolution produces a local receptive field, which may struggle to capture long-range spatio-temporal dependencies effectively. To compensate for such shortage, this paper introduces a novel approach called the Transformer-based Multi-level Attention Integration Network (TMAI-Net) for video saliency prediction. TMAI-Net is designed as a two-stream encoder-decoder model, carefully integrating multi-level features of semantic information. Our model incorporates a Multi-level Interactive Attention(MLIA) module and a Transformer, both implemented based on self-attention mechanism, which are placed at different levels of the model to capture long-range spatio-temporal feature dependencies. Additionally, our model operates on input video frames and attentional patches, allowing the Transformer module to capture structural similarities between related objects in global features and attention features. This, in turn, enables the model to allocate increased attention to salient areas. The efficacy of our proposed approach is validated through extensive experiments conducted on three widely recognized benchmark datasets. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

WFANet-DDCL: Wavelet-Based Frequency Attention Network and Dual Domain Consistency Learning for 7T MRI Synthesis From 3T MRI

引用

IEEE Transactions on Circuits and Systems for Video technology 2025年第6期35卷 5617-5632页

作者： Liu, Xiaolong Qiu, Song Zhou, Mei Le, Weijie Li, Qingli Wang, Yan East China Normal University Shanghai Key Laboratory of Multidimensional Information Processing MoE Engineering Research Center of Software/Hardware Co-Design Technology and Application Shanghai200241 China Shanghai Jiao Tong University School of Medicine Radiology Department Ninth People’s Hospital Shanghai200011 China

Ultra-high field magnetic resonance imaging (MRI), such as 7-Tesla (7T) MRI, provides significantly enhanced tissue contrast and anatomical details compared to 3T MRI. However, 7T MRI scanners are more costly and less accessible in clinical settings than 3T scanners. In this paper, we propose a wavelet-based frequency attention network (WFANet) and a semi-supervised method named dual domain consistency learning (DDCL), and combine them to form a WFANet-DDCL framework for 7T MRI synthesis. WFANet leverages the frequency sensitivity of the proposed wavelet-based frequency attention encoder (WFAE) along with the large receptive field of dilated convolution. WFAE is proposed as an independent module to capture multi-scale frequency attention via the proposed wavelet-based frequency attention (WFA) mechanism. WFAE can be integrated into any backbone network as a plug-and-play component and improve network performance. To tackle the challenge of limited paired data for network training, DDCL is proposed to take advantage of both paired and unpaired data. Frequency domain perturbation is proposed and combined with Gaussian noise to regularize the supervised learning process in dual domains, better avoiding overfitting. Extensive experimental results demonstrate that WFANet-DDCL can achieve comparable performance to state-of-the-art supervised methods even using 66% of all paired data. © 1991-2012 IEEE.

关键词： Semi-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Improving the Performance of Lattice Boltzmann Method with Pipelined Algorithm on A Heterogeneous Multi-zone Processor 23rd

Improving the Performance of Lattice Boltzmann Method with...

引用

23rd International Conference on Parallel and Distributed Computing, Applications, and Technologies, PDCAT 2022

作者： Zhang, Qingyang Xu, Lei Chen, Rongliang Chen, Lin Chen, Xinhai Wang, Qinglin Liu, Jie Yang, Bo Science and Technology on Parallel and Distributed Processing Laboratory National University of Defense Technology Changsha410000 China Laboratory of Software Engineering for Complex Systems National University of Defense Technology Changsha410000 China Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences Shenzhen518055 China

ISBN: (纸本)9783031299261

Lattice Boltzmann method (LBM) has become a powerful method in computational fluid dynamics and has drawn more and more attention in high-performance computing due to its particulate nature and local dynamics, especially on recent multi-core or many-core platforms. This paper develops a parallel software framework for 3D LBM simulation on a heterogeneous multi-zone processor, MT-3000. An improved pipelined algorithm named Pencil-H is proposed, which can not only fully exploit the advantages of each component of MT-3000 but also overlap the time of calculation and communication. Moreover, an architecture-aware multi-level parallelization algorithm is developed to fully utilize the computational performance of MT-3000. A benchmark test is performed to verify the reliability and test the performance of the LBM code. Experimental results show that the optimized code achieves a 32.02 × speedup compared with using 16 CPU cores and achieves a performance of 286.03MLUPS which reaches 72.3% of the theoretical peak performance. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Computational fluid dynamics

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：