检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Jiang, Yanfeng Sun, Ning Xie, Xueshuo Yang, Fei Li, Tao College of Computer Science Nankai University Tianjin China Tianjin Key Laboratory of Network and Data Security Technology Tianjin China Zhejiang Lab Zhejiang Hangzhou China Haihe Lab of ITAI Tianjin China

Vision Transformers (ViTs) have exhibited exceptional performance across diverse computer vision tasks, while their substantial parameter size incurs significantly increased memory and computational demands, impeding effective inference on resource-constrained devices. Quantization has emerged as a promising solution to mitigate these challenges, yet existing methods still suffer from significant accuracy loss at low-bit. We attribute this issue to the distinctive distributions of post-LayerNorm and post-GELU activations within ViTs, rendering conventional hardware-friendly quantizers ineffective, particularly in low-bit scenarios. To address this issue, we propose a novel framework called Activation-Distribution-Friendly post-training Quantization for Vision Transformers, ADFQ-ViT. Concretely, we introduce the Per-Patch Outlier-aware Quantizer to tackle irregular outliers in post-LayerNorm activations. This quantizer refines the granularity of the uniform quantizer to a per-patch level while retaining a minimal subset of values exceeding a threshold at full-precision. To handle the non-uniform distributions of post-GELU activations between positive and negative regions, we design the Shift-Log2 Quantizer, which shifts all elements to the positive region and then applies log2 quantization. Moreover, we present the Attention-score enhanced Module-wise Optimization which adjusts the parameters of each quantizer by reconstructing errors to further mitigate quantization error. Extensive experiments demonstrate ADFQ-ViT provides significant improvements over various baselines in image classification, object detection, and instance segmentation tasks at 4-bit. Specifically, when quantizing the ViT-B model to 4-bit, we achieve a 10.23% improvement in Top-1 accuracy on the ImageNet dataset. Copyright © 2024, The Authors. All rights reserved.

关键词： Distribution transformers

来源：评论

学校读者我要写书评

暂无评论

Taxi Station Location Model Based on Spatio-Temporal Demand Cube

Taxi Station Location Model Based on Spatio-Temporal Demand ...

引用

International Conference on networking and network Applications (NaNA)

作者： LuLu Gao XiaoNan Li ZhiCheng Wei Key Lab of Network and Information Security Shijiazhuang Hebei Province China College of Computer and Cyber Security Shijiazhuang Hebei Province China Hebei Provincial Engineering Research Center for Supply Chain Big Data Analytics & Security Hebei Normal University Shijiazhuang Hebei Province China

With the widespread use of mobile apps, empty taxis can easily receive new orders without having to roam around. Therefore, it is necessary to plan taxi stations where empty taxis can park. In the location selection problem, the spatial and temporal distribution of passenger demand needs to be considered to ensure that the selected location can meet the actual needs of passengers. In this paper, a Spatio-Temporal Cube Demand Coverage Model (STC-DCM) is proposed for taxi station location selection. Firstly, a spatiotemporal demand cube is constructed to obtain fine-grained user demands based on time and space, and genetic algorithm is used to maximize the demand coverage rate. Then, the final location selection results are obtained using the fusion module and adjustment optimization module, where the fusion module integrates multiple location selection solutions using ensemble learning method. Experimental results show that our proposed model can better cover dynamic passenger demand compared to traditional models.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Adfq-Vit: Activation-Distribution-Friendly Post-Training Quantization for Vision Transformers

SSRN

引用

SSRN 2024年

关键词： Chemical activation

来源：评论

学校读者我要写书评

暂无评论

CustomFair: A Customized Fairness Method for Federated Recommender Systems in Social Internet of Things

引用

IEEE Internet of Things Journal 2024年

作者： Chen, Guorong Li, Chao Du, Fei Yuan, Xiaohan Chi, Cheng Yin, Zihang Wang, Bin Li, Tao Bao, Xuhua Wang, Wei Beijing Jiaotong University Beijing Key Laboratory of Security and Privacy in Intelligent Transportation Beijing100044 China China Academy of Information and Communications Technology Beijing China Network and Data Security Hangzhou310053 China Nankai University College of Computer Science Tianjin300350 China LTD Sangfor Technologies Co Beijing China Xi'an Jiaotong University Ministry of Education Key Lab for Intelligent Networks and Network Security Xi'an China

In the Social Internet of Things (SIoT), edge computing integrates artificial intelligence to learn intricate relationships. The scale and complexity of SIoT cause a data explosion from diverse objects, hindering tailored services to users who own objects. Moreover, conventional edge computing in SIoT depends on centralized data collection, raising concerns about data privacy. To address the above two issues, federated recommender systems (FRSs) present a promising solution. FRSs can provide SIoT services to users and train a shared model while retaining sensitive data locally on objects. However, as FRSs are driven by data, they are inherently susceptible to algorithmic bias, raising substantial fairness concerns that have attracted considerable attention in SIoT. Recent fairness studies predominantly concentrate on a single sensitive attribute for users, thereby overlooking their autonomy. Therefore, we propose CustomFair, a personalized fairness framework that enables users in FRSs to select preferred sensitive attributes and acquire satisfied recommendation services in SIoT scenarios. First, we define customized fairness to ensure group fairness based on users' sensitive attributes. The server segments users into subgroups in a privacy-preserving manner. Second, CustomFair employs the DynBalance method with a flexible regularization coefficient to improve recommendation performance and utilizes the AdaptEpoch strategy to achieve fairness. Extensive experiments indicate that CustomFair improves recommendation performance by 0.1 ∼ 42.92 and enhances fairness by reducing disparities of 0.03 ∼ 5.41 compared to two baselines across three datasets. © 2014 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

HTtext: A TextCNN-based pre-silicon detection for hardware Trojans

HTtext: A TextCNN-based pre-silicon detection for hardware T...

引用

IEEE International Conference on Big data and Cloud Computing (BdCloud)

作者： Yi Xu Zhenyi Chen Binhong Huang Ximeng Liu Chen Dong College of Computer and Data Science Fuzhou University Fuzhou China University of South Floride Tampa USA Key Lab of Information Security of Network Systems (Fuzhou University) Fujian Key Laboratory of Network Computing and Intelligent Information Processing (Fuzhou University)

With the advent of the intelligence era, the usage and investment for chips have risen year by year, causing the demand for chip security. Machine learning (ML) analysis has made progress as a pre-silicon hardware Trojan (HT) detection technology. Whereas the performance of existing methods almost relies on the accuracy of multi-feature representation, moreover, it is difficult to extract features manually and easily cause unstable classifier performance (namely uncertainty). In this paper, an automatic feature extraction detecting model is first proposed, named HTtext, which generates simple path sentences from chip netlists and employs TextCNN, a deep learning algorithm, to distinguish HT circuits. The pre-training for TextCNN only uses the automatic single-feature calculation to avoid the uncertainty problem. Additionally, the model can obtain non-repetitive HT component information expression, which satisfies the stable detection performance. To measure the efficiency and balance of the model, the paper proposes the concept of the Stability Efficiency Index (SEI). In the experimental results for the benchmark netlists, not only the average accuracy (ACC) in TextCNN is as high as 99.26%, but also its SEI value ranks first in all comparison classifiers, which proves that the proposed HTtext model has high stability in generality.

关键词： Ports (computers) Semiconductor device measurement Uncertainty Stability criteria Feature extraction Hardware Circuit stability

来源：评论

学校读者我要写书评

暂无评论

A Dynamic Demand-driven Smart Manufacturing for Mass Individualization Production

A Dynamic Demand-driven Smart Manufacturing for Mass Individ...

引用

IEEE International Conference on Systems, Man and Cybernetics

作者： Qiyu Hong Zhenyi Chen Chen Dong Qiancheng Xiong College of Computer and Data Science Fuzhou University Fuzhou China Key Lab of Information Security of Network Systems Fuzhou University Fujian Key Laboratory of Network Computing and Intelligent Information Processing Fuzhou University University of South Florida Tampa United States

ISBN: (纸本)9781665442084

Industry 4.0 intends to realize mass individualization production driven by customers’ demands, thus generating several advanced manufacturing modes. However, the existing modes have disadvantages such as low customer participation, low response for modifications in real-time, and low utilization of distributed idle productivity resources. The above deficiencies would not cater to the idea of Industry 4.0 well, making the cost of individualization products too high or the degree of individualization too low. In this paper, a customer highly involved dynamic demand-driven smart manufacturing mode and a mathematical model are presented for mass individualization production. Under this mode, customers can be prosumers, participating in the whole production process and proposing modifications. Meanwhile, multiple distributed idle productivity resources are employed with dynamic demand-driven manufacturing. In consequence of the mathematical model and algorithms, the multiple task allocation among multiple factories problem is solved, maximizing the total profit of available factories with high efficiency and low resource waste. Experiments and comparisons are made about the feasibility and efficiency of the algorithms.

关键词： Productivity Heuristic algorithms Dynamic scheduling Real-time systems Production facilities Mathematical models Fourth Industrial Revolution

来源：评论

学校读者我要写书评

暂无评论

Contrastive Multi-view Hyperbolic Hierarchical Clustering

arXiv

引用

arXiv 2022年

作者： Lin, Fangfei Bai, Bing Bai, Kun Ren, Yazhou Zhao, Peng Xu, Zenglin University of Electronic Science and Technology of China Chengdu China Tencent Security Big Data Lab Tencent Inc. China Harbin Institute of Technology Shenzhen China Department of Network Intelligence Peng Cheng National Lab Shenzhen China

Hierarchical clustering recursively partitions data at an increasingly finer granularity. In real-world applications, multi-view data have become increasingly important. This raises a less investigated problem, i.e., multi-view hierarchical clustering, to better understand the hierarchical structure of multi-view data. To this end, we propose a novel neural network-based model, namely Contrastive Multi-view Hyperbolic Hierarchical Clustering (CMHHC). It consists of three components, i.e., multi-view alignment learning, aligned feature similarity learning, and continuous hyperbolic hierarchical clustering. First, we align sample-level representations across multiple views in a contrastive way to capture the view-invariance information. Next, we utilize both the manifold and Euclidean similarities to improve the metric property. Then, we embed the representations into a hyperbolic space and optimize the hyperbolic embeddings via a continuous relaxation of hierarchical clustering loss. Finally, a binary clustering tree is decoded from optimized hyperbolic embeddings. Experimental results on five real-world datasets demonstrate the effectiveness of the proposed method and its components. Copyright © 2022, The Authors. All rights reserved.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

Secure Deduplication Against Frequency Analysis Attacks

Secure Deduplication Against Frequency Analysis Attacks

引用

International Conference on Mobile Ad-hoc and Sensor networks, MSN

作者： Hang Chen Guanxiong Ha Yuchen Chen Haoyu Ma Chunfu Jia College of Cyber Science Nankai University Tianjin China Tianjin Key Laboratory of Network and Data Security Technology Tianjin China Zhejiang Lab Hangzhou China The School of Cyber Engineering Xidian University Xi’an China

Message-locked Encryption (MLE) is the most common approach used in encrypted deduplication systems. However, the systems based on MLE are vulnerable to frequency analysis attacks, because MLE encrypts the identical plain texts into the identical ciphertexts, which is deterministic. The state-of-the- art defense scheme, which named TED, lacks key verification and uses a single key server to record frequency information. Once the key server is compromised, TED will be vulnerable to brute-force attacks. In addition, TED's key generation algorithm needs to be designed more exquisitely to strengthen protection, and its security indicator is not comprehensive. We propose SDAF, which supports key verification and enhanced protection against frequency analysis attacks. Based on chameleon hash, SDAF realizes key verification to prevent malicious key servers from generating fake encryption keys. In order to disturb the frequency information, SDAF introduces reservoir sample to generate uniformly distributed encryption keys, and uses multiple key servers, which interact with each other via multi-party PSI and rotate spontaneously to avoid the single point of failure. Moreover, a new indicator Kurtosis is pointed out to evaluate the security against frequency analysis attacks. We implement the prototypes of SDAF. The experiments of the real-world data sets show that, compared with the existing schemes, SDAF can better resist frequency analysis attacks with lower time overheads.

关键词： Maximum likelihood estimation Time-frequency analysis data privacy Art Prototypes Resists Reservoirs

来源：评论

学校读者我要写书评

暂无评论

Deep Feature Surgery: Towards Accurate and Efficient Multi-Exit networks

arXiv

引用

arXiv 2024年

作者： Gong, Cheng Chen, Yao Luo, Qiuyang Lu, Ye Li, Tao Zhang, Yuzhi Sun, Yufei Zhang, Le College of Software Nankai University China College of Computer Science Nankai University China National University of Singapore Singapore School of Information and Communication Engineering University of Electronic Science and Technology of China China HAIHE Lab of ITAI China Tianjin Key Laboratory of Network and Data Security Technology China Key Laboratory of Data and Intelligent System Security Ministry of Education

Multi-exit network is a promising architecture for efficient model inference by sharing backbone networks and weights among multiple exits. However, the gradient conflict of the shared weights results in sub-optimal accuracy. This paper introduces Deep Feature Surgery (DFS), which consists of feature partitioning and feature referencing approaches to resolve gradient conflict issues during the training of multi-exit networks. The feature partitioning separates shared features along the depth axis among all exits to alleviate gradient conflict while simultaneously promoting joint optimization for each exit. Subsequently, feature referencing enhances multi-scale features for distinct exits across varying depths to improve the model accuracy. Furthermore, DFS reduces the training operations with the reduced complexity of backpropagation. Experimental results on Cifar100 and ImageNet datasets exhibit that DFS provides up to a 50.00% reduction in training time and attains up to a 6.94% enhancement in accuracy when contrasted with baseline methods across diverse models and tasks. Budgeted batch classification evaluation on MSDNet demonstrates that DFS uses about 2× fewer average FLOPs per image to achieve the same classification accuracy as baseline methods on Cifar100. The code is available at https: //***/GongCheng1919/dfs. Copyright © 2024, The Authors. All rights reserved.

关键词： Budget control

来源：评论

学校读者我要写书评

暂无评论

Constructing a Science of Cyber-Resilience for Military Systems

Constructing a Science of Cyber-Resilience for Military Syst...

引用

2017 NATO IST-153/RWS-21 Workshop on Cyber Resilience, IST-153 2017

作者： Alexeev, Alexander Henshel, Diane S. Levitt, Karl McDaniel, Patrick Rivera, Brian Templeton, Steven Weisman, Mike Data Analysis and Modeling Indiana University School of Public and Environmental Affairs 1315 E 10th Street BlooomingtonIN47405 Indiana University School of Public and Environmental Affairs 1315 E 10th Street BloomingtonIN47405 Computer Science College of Engineering University of California Davis 1 Shields Ave. DavisCA05616 School of Electrical Engineering and Computer Science Pennsylvania State University Information Sciences and Technology Building University ParkPA16802-6823 Network Science Division Tactical Network Assurance Branch U.S. Army Research Laboratory 2800 Powder Mill Road AdelphiMD20783 Computer Security Research Lab Department of Computer Science University of California Davis One Shields Avenue DavisCA95616 Network Science Division Network Security Branch U.S. Army Research Laboratory ICF International 2800 Powder Mill Road AdelphiMD20783

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：