检索结果-内蒙古大学图书馆

IEEE Workshop on Signal Processing Advances in Wireless Communications (SPAWC)

作者： Junshen Su Xijun Wang Xiang Chen Ray-Guang Cheng School of Electronics and Information Technology Sun Yat-Sen University Guangzhou China Guangdong Provincial Key Laboratory of Big Data Computing The Chinese University of Hong Kong Shenzhen China Department of Electronic and Computer Engineering National Taiwan University of Science and Technology Taipei Taiwan

Communication efficiency is one of the key bottlenecks in Federated Learning (FL). Compression techniques, such as sparsification and quantization, are used to reduce communication overhead. However, joint designs of these techniques under communication constraints are not well-explored. This paper investigates the joint uplink compression problem in communication-constrained FL systems. We propose a Block-TopK sparsification scheme to reduce the proportion of bits used for locating entries of a sparsified vector. Considering the communication constraints, an optimization formulation is proposed to minimize the compression error. By solving the optimization problem, our joint compression method provides a better trade-off between sparsity budget and bit width. Numerical results demonstrate that our approach achieves 99.96% of baseline accuracy with only 1.56% of the baseline communication overhead when training ResNet-18 on CIFAR-10.

关键词：

来源：评论

学校读者我要写书评

暂无评论

An Efficient data Prefetch Strategy for Deep Learning Based on Non-volatile Memory 15th

An Efficient Data Prefetch Strategy for Deep Learning Based ...

引用

15th International Conference on Green, Pervasive, and Cloud computing, GPC 2020

作者： Jiang, Wenbin Liu, Pai Jin, Hai Peng, Jing National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology Huazhong University of Science and Technology Wuhan430074 China

ISBN: (纸本)9783030642426

Deep learning (DL) systems usually utilize asynchronous prefetch to improve data reading performance. However, the efficiency of the data transfer path from hard disk to DRAM is still limited by disk performance. The emerging non-volatile memory (NVRAM) provides a novel solution for this problem, while few existing researches have considered it. We propose a novel efficient data prefetch strategy for DL based on a heterogeneous memory system combining NVRAM with DRAM. Benefitting from the large capacity and fast reading speed of NVRAM, the strategy uses an asynchronous reading method named sliding NVRAM cache (SNC) to improve the performance of the data transfer paths. A sliding window is applied to map the data from disk to NVRAM and continuously update the data, while non-ideal writing performance of NVRAM can be remitted to a large extent in this strategy. Experiments show that SNC can improve the time performance of diverse deep neural networks training by more than 30%. © 2020, Springer Nature Switzerland AG.

关键词： Dynamic random access storage

来源：评论

学校读者我要写书评

暂无评论

Generalized Face Forgery Detection via Adaptive Learning for Pre-trained Vision Transformer

arXiv

引用

arXiv 2023年

作者： Luo, Anwei Cai, Rizhao Kong, Chenqi Ju, Yakun Kang, Xiangui Huang, Jiwu Kot, Alex C. The School of Information Technology Jiangxi University of Finance and Economics Nanchang330013 China The School of Computer Science and Engineering Sun Yat-Sen University Guangzhou510006 China Lab. School of Electrical and Electronic Engineering Nanyang Technology University Singapore The Guangdong Key Laboratory of Intelligent Information Processing National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Shenzhen518060 China The China-Singapore International Joint Research Institute Singapore

With the rapid progress of generative models, the current challenge in face forgery detection is how to effectively detect realistic manipulated faces from different unseen domains. Though previous studies show that pre-trained Vision Transformer (ViT) based models can achieve some promising results after fully fine-tuning on the Deepfake dataset, their generalization performances are still unsatisfactory. One possible reason is that fully fine-tuned ViT-based models may disrupt the pre-trained features [1], [2] and overfit to some data-specific patterns [3]. To alleviate this issue, we present a Forgery-aware Adaptive Vision Transformer (FA-ViT) under the adaptive learning paradigm, where the parameters in the pre-trained ViT are kept fixed while the designed adaptive modules are optimized to capture forgery features. Specifically, a global adaptive module is designed to model long-range interactions among input tokens, which takes advantage of self-attention mechanism to mine global forgery clues. To further explore essential local forgery clues, a local adaptive module is proposed to expose local inconsistencies by enhancing the local contextual association. In addition, we introduce a fine-grained adaptive learning module that emphasizes the common compact representation of genuine faces through relationship learning in fine-grained pairs, driving these proposed adaptive modules to be aware of fine-grained forgery-aware information. Extensive experiments demonstrate that our FA-ViT achieves state-of-the-arts results in the cross-dataset evaluation, and enhances the robustness against unseen perturbations. Particularly, FA-ViT achieves 93.83% and 78.32% AUC scores on Celeb-DF and DFDC datasets in the cross-dataset evaluation. The code and trained model have been released at: https://***/LoveSiameseCat/FAViT. Copyright © 2023, The Authors. All rights reserved.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

BED: A Block-Level Deduplication-Based Container Deployment Framework 15th

BED: A Block-Level Deduplication-Based Container Deployment ...

引用

15th International Conference on Green, Pervasive, and Cloud computing, GPC 2020

作者： Zhang, Shiqiang Wu, Song Fan, Hao Zou, Deqing Jin, Hai National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology Huazhong University of Scinece and Technology Wuhan430074 China

ISBN: (纸本)9783030642426

Container technology has gained great popularity in cloud environment since containers provide near-native performance and are lighter and less expensive than traditional virtual machines. However, starting up a non-local container, whose image is unavailable locally, is time-consuming. The reason is that pulling an image from a remote registry requires a long time. In this paper, we experimentally find that a lot of redundant data exists among image layers when we pull images from the registry. This redundant data causes additional pull time and makes the startup of a non-local container slower. To minimize the amount of pulled data while providing intact images, we propose a container deployment framework, BED, based on block-level deduplication. To be specific, BED stores an image layer as numerous data blocks and a fingerprint list which is generated based on these data blocks. When BED needs to pull an image, it pulls the fingerprint lists of the image, deduplicates these fingerprint lists, and pulls the data blocks non-existing locally from the registry. Based on the local and pulled blocks, BED reconstructs the image layers for a container. Experiments show that compared with original Docker, BED reduces the time of pulling images by 35% on average and saves about 48% data transmission in the network. © 2020, Springer Nature Switzerland AG.

关键词： Containers

来源：评论

学校读者我要写书评

暂无评论

Dynamic Searchable Scheme with Forward Privacy for Encrypted Document Similarity

Dynamic Searchable Scheme with Forward Privacy for Encrypted...

引用

IEEE International Conference on Trust, Security and Privacy in computing and Communications (TrustCom)

作者： Mustafa A. Al Sibahee Chengwen Luo Jin Zhang Yijing Huang Zaid Ameen Abduljabbar National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Shenzhen PR China College of Computer Science and Software Engineering Shenzhen University Shenzhen PR China Computer Technology Engineering Department Iraq University College Basrah Iraq Department of Computer Science College of Education for Pure Sciences University of Basrah Basrah Iraq Shenzhen Institute Huazhong University of Science and Technology Shenzhen PR China

ISBN: (数字)9798350381993

ISBN: (纸本)9798350382006

Document retrieval plays an essential role in many real-world applications especially when the data storage is outsourced. Due to the great advantages offered by cloud computing, clients tend to outsource their personal data to remote servers maintained by external service providers. This raises serious privacy concerns about outsourced data because such providers are usually considered untrusted entities. The majority of previous schemes of document similarity search share the same limitation: they focus mainly on static collections. Dynamic searchable schemes (DSE) allow adding or removing documents at the expense of more leakage than static schemes. To thwart certain attacks, DSE schemes should support forward privacy property, which ensures that newly added documents cannot be related to previously issued search queries. We design and implement dynamic secure similarity search schemes with forward privacy for textual documents utilizing simhash method for hamming similarity. Our scheme provides an efficient search time and a sufficient level of privacy. To show the practicality of our proposed scheme, we performed excremental results with large document collections.

关键词： Privacy data privacy Cloud computing Memory Search problems Cryptography Servers

来源：评论

学校读者我要写书评

暂无评论

Differential Privacy-based Avatar Update in Cooperative Intelligent Transportation system Metaverse

Differential Privacy-based Avatar Update in Cooperative Inte...

引用

International Conference on Computer and Communications (ICCC)

作者： Salabat Khan Fei Luo Mansoor Khan Insaf Ullah Farhan Amin Shamsher Ullah Mussadiq Abdul Rahim School of Computer and Enformation Engineering Qilu Institute of Technology Jinan China School of Computing and Information Technology Great Bay University Dongguan China School of Intelligent Manufacturing and Control Engineering Qilu Institute of Technology Jinan China Institute for Analytics and Data Science University of Essex Colchester UK School of Computer Science and Engineering Yeungnam University South Korea National Engineering Laboratory for Big Data Shenzhen University Shenzhen China King Fahd University of Petroleum and Minerals (KFUPM) Dhahran Saudi Arabia

ISBN: (数字)9798331507077

ISBN: (纸本)9798331507084

Metaverse is the next generation and the successor to the Internet. Various sectors allocated capital to adapt the metaverse due to its inherent importance. Cooperative Intelligent Transportation systems (C-ITS) is one of the important sectors that can benefit from metaverse adaptation and integration. The C-ITS and metaverse integration give rise to new use cases and cyber attacks. However, traditional Pseudonym Certificate (PCert) authentication based on Vehicular Public Key Infrastructure (VPKI) cannot meet the new use cases and thwart cyber attacks. We present a Differential Privacy (DP)-based PCert and avatar update model to ensure the privacy of end-devices (e.g., vehicles) in the C-ITS metaverse. The proposed DP-based model adds noise to avatar attributes (e.g., color, behavior) to prevent avatar linkage during the PCert update process. Our method ensures that PCerts and avatars cannot be linked to end-devices and the previous PCerts and avatars. We simulate the DP-based method to investigate it in terms of privacy, latency, color distortion, and adversarial success rate, and the results show that DP-based is effective and practical for avatar updates.

关键词： Privacy Metaverse Avatars Computational modeling Authentication Public key Color Colored noise Intelligent transportation systems Cyberattack

来源：评论

学校读者我要写书评

暂无评论

Hardware-Supported Remote Persistence for Distributed Persistent Memory

Hardware-Supported Remote Persistence for Distributed Persis...

引用

Supercomputing Conference

作者： Zhuohui Duan Haodi Lu Haikun Liu Xiaofei Liao Hai Jin Yu Zhang Song Wu National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computing Science and Technology Huazhong University of Science and Technology Wuhan China

ISBN: (数字)9781450384421

ISBN: (纸本)9781665483902

The advent of Persistent Memory (PM) necessitates an evolution of Remote Direct Memory Access (RDMA) technologies for supporting remote data persistence. Previous software-based solutions require remote CPU intervention and postpone the visibility of remote persistence. In this paper, we design several hardware-supported RDMA primitives to flush data from the volatile cache of RDMA Network Interface Cards (RNICs) to the PM. We also propose durable RPCs based on the proposed RDMA Flush primitives to support remote data persistence and fast failure recovery. We emulate the performance of RDMA Flush primitives through other RDMA prim-itives, and compare our proposals with several state-of-the-art RPCs in a real testbed equipped with PM and InfiniBand networks. Experimental results show that our proposals can improve the throughput of RPCs by up to 90%, and reduce the 99th percentile latency by up to 49%. The experimental studies also provide instructive guidelines for designing RDMA-based distributed PM systems.

关键词： High performance computing Throughput Proposals Network interfaces Guidelines

来源：评论

学校读者我要写书评

暂无评论

Clustering with adaptive graph learning and spectral rotation

Clustering with adaptive graph learning and spectral rotatio...

引用

7th International Conference on Vision, Image and Signal Processing (ICVISP 2023)

作者： G. Yuan J. Zhou C. Huang C. Gao Y. Wang X. Shen National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Shenzhen 518060. Guangdong People's Republic of China College of Computer Science and Software Engineering Shenzhen University Shenzhen 518060 Guangdong People's Republic of China College of Mathematics and Statistics Shenzhen University Shenzhen 518060 Guangdong People's Republic of China

Prototype-based clustering algorithms have garnered considerable attention in the field of machine learning due to their efficiency and interpretability. Nonetheless, these algorithms often face performance degradation when confronted with high-dimensional or non-ellipsoidal data distributions. To surmount these challenges, this study introduces a novel clustering approach, dubbed Clustering with Adaptive Graph learning and Spectral Rotation (CAGSR). In CAGSR, the imposed spectral rotation operation mitigates the discrepancy between the membership matrix, which adheres to the notion of fuzzy clustering, and the spectral representations derived from an adaptive graph rather than a predefined one. This enables the generation of a comprehensive representation of the data across multiple spaces. Furthermore, the clustering and graph learning tasks are jointly optimized in a projected subspace, which can effectively reduce the adverse impact caused by irrelevant features in the original space. The proposed method seamlessly integrates fuzzy clustering, graph structure learning, and spectral rotation into a unified model, facilitating the detection of intrinsic structures. Experimental evaluations conducted on benchmark data sets substantiate the effectiveness of CAGSR when compared to related clustering approaches.

关键词：

来源：评论

学校读者我要写书评

暂无评论

SDGFormer: An Efficient Convolution Network Structurally Similar to Transformer

SDGFormer: An Efficient Convolution Network Structurally Sim...

引用

IEEE International Conference on Multimedia and Expo (ICME)

作者： Chaohao Wen Xun Gong School of Computing and Artificial Intelligence SWJTU Chengdu China Engineering Research Center of Sustainable Urban Intelligent Transportation Ministry of Education China National Engineering Laboratory of Integrated Transportation Big Data Application Technology SWJTU Chengdu China Manufacturing Industry Chains Collaboration and Information Support Technology Key Laboratory of Sichuan Province SWJTU Chengdu China

Deep Neural Networks (DNN) have achieved extraordinary success in many visual recognition tasks. Visual Transformer (ViT), which is derived from Natural Language Processing (NLP), has achieved state-of-the-art (SOTA) results on many tasks due to its capability of capturing long-range dependencies in visual data. However, Existing ViT models are challenging to deploy on devices due to their massive computational consumption, huge memory overhead, and reliance on large datasets. In this work, we address these issues by replacing some computationally expensive and memory-intensive modules in ViT with standard Convolutional Neural Network (CNN) modules. Firstly, we propose an efficient Self-Attention module called SDG-Attention (SDGA) with linear space and time complexity, and an economical FeedForward Network (FFN) composed of group convolution and shuffle channel (SFFN). Then, we develop a lightweight CNN model with SDGA and SFFN, SDGFormer, which embraces several priors of ViT and is LayerNorm-Free. We evaluate SDGFormer on ImageNet-1K and Mini-ImageNet, and the SDGFormer-S achieves a comparable top-1 accuracy of 77.6% on ImageNet-1K with 9.1M parameters and 1.6 GFlops regimes. Moreover, our SDGFormer-T achieves SOTA performance on Mini-ImageNet with 83.3% accuracy, demonstrating good generalization on small datasets without extra data.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Continual Test-Time Adaptation for Single Image Defocus Deblurring via Causal Siamese Networks

arXiv

引用

arXiv 2025年

作者： Cui, Shuang Li, Yi Li, Jiangmeng Tang, Xiongxin Su, Bing Xu, Fanjiang Xiong, Hui National Key Laboratory of Space Integrated Information System Institute of Software Chinese Academy of Sciences University of Chinese Academy of Sciences China Beijing Key Laboratory of Big Data Management and Analysis Methods Gaoling School of Artificial Intelligence Renmin University of China China Department of Computer Science & Engineering the Hong Kong University of Science and Technology Hong Kong

Single image defocus deblurring (SIDD) aims to restore an all-in-focus image from a defocused one. Distribution shifts in defocused images generally lead to performance degradation of existing methods during out-of-distribution inferences. In this work, we gauge the intrinsic reason behind the performance degradation, which is identified as the heterogeneity of lens-specific point spread functions. Empirical evidence supports this finding, motivating us to employ a continual test-time adaptation (CTTA) paradigm for SIDD. However, traditional CTTA methods, which primarily rely on entropy minimization, cannot sufficiently explore task-dependent information for pixel-level regression tasks like SIDD. To address this issue, we propose a novel Siamese networks-based continual test-time adaptation framework, which adapts source models to continuously changing target domains only requiring unlabeled target data in an online manner. To further mitigate semantically erroneous textures introduced by source SIDD models under severe degradation, we revisit the learning paradigm through a structural causal model and propose Causal Siamese networks (CauSiam). Our method leverages large-scale pre-trained vision-language models to derive discriminative universal semantic priors and integrates these priors into Siamese networks, ensuring causal identifiability between blurry inputs and restored images. Extensive experiments demonstrate that CauSiam effectively improves the generalization performance of existing SIDD methods in continuously changing domains. Copyright © 2025, The Authors. All rights reserved.

关键词： Semantics

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：