检索结果-内蒙古大学图书馆

作者： Yuan, Xu Shen, Xiangjun Mehta, Sumet Li, Teng Ge, Shiming Zha, Zhengjun School of Computer Science and Telecommunication Engineering Jiangsu University Jiangsu Zhenjiang212013 China School of electrical engineering and Automation Anhui University Jiangsu Anhui Hefei230039 China Institute of information engineering CAS Chinese academy of sciences Beijing100086 China School of data science University of science and technology of china Anhui Hefei230026 China

Weight normalization (WN) can help to stabilize the distribution of activations over layers, which boost the performance of DNNs in generalization. In this paper, we further propose deep structural weight normalization (DSWN) methods to inject the network structure measurements into the WN to fully acknowledge the data propagation through the neural network. In DSWN, two novel structural measurements are developed to impose regularity on each network weight using different penalty matrices. One is sparsity measurement (DSWN-SM). In this measurement, L1,2 weight regularization is applied in our proposed model to promote competition for features between network weights to obtain a sparsity network and finally prune the network. The other is neuron measurement (DSWN-NM). It uses L2 norm of column weight to scale up or down the importance of each intermediate neuron, which leads to accelerating the speed of network convergence. Extensive experiments on several benchmark image datasets using fully connected network and convolution neural network are performed, and the proposed DSWN-SM and DSWN-NM methods are compared with state-of-the-art sparsity and weight normalization methods. The results show that DSWN-SM can reduce the number of trainable parameters while guaranteeing high accuracy, whereas DSWN-NM can accelerate the convergence while improving the performance of deep networks. © 2021, The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： Neurons

来源：评论

学校读者我要写书评

暂无评论

Unlearnable 3D point clouds: class-wise transformation is all you need 24

Unlearnable 3D point clouds: class-wise transformation is al...

引用

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Xianlong Wang Minghui Li Wei Liu Hangtao Zhang Shengshan Hu Yechao Zhang Ziqi Zhou Hai Jin National Engineering Research Center for Big Data Technology and System and Services Computing Technology and System Lab and Hubei Engineering Research Center on Big Data Security and Hubei Key Laboratory of Distributed System Security and School of Cyber Science and Engineering Huazhong University of Science and Technology School of Software Engineering Huazhong University of Science and Technology Hubei Engineering Research Center on Big Data Security and Hubei Key Laboratory of Distributed System Security and School of Cyber Science and Engineering Huazhong University of Science and Technology National Engineering Research Center for Big Data Technology and System and Services Computing Technology and System Lab and Cluster and Grid Computing Lab and School of Computer Science and Technology Huazhong University of Science and Technology

ISBN: (纸本)9798331314385

Traditional unlearnable strategies have been proposed to prevent unauthorized users from training on the 2D image data. With more 3D point cloud data containing sensitivity information, unauthorized usage of this new type data has also become a serious concern. To address this, we propose the first integral unlearnable framework for 3D point clouds including two processes: (i) we propose an unlearnable data protection scheme, involving a class-wise setting established by a category-adaptive allocation strategy and multi-transformations assigned to samples; (ii) we propose a data restoration scheme that utilizes class-wise inverse matrix transformation, thus enabling authorized-only training for unlearnable data. This restoration process is a practical issue overlooked in most existing unlearnable literature, i.e., even authorized users struggle to gain knowledge from 3D unlearnable data. Both theoretical and empirical results (including 6 datasets, 16 models, and 2 tasks) demonstrate the effectiveness of our proposed unlearnable framework. Our code is available at https://***/CGCL-codes/UnlearnablePC.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Efficient Modeling Attack on Multiplexer PUFs via Kronecker Matrix Multiplication

引用

IEEE Transactions on computer-Aided Design of Integrated Circuits and Systems 2025年

作者： Wang, Hongfei Wan, Caixue Jin, Hai Huazhong University of Science and Technology National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Hubei Key Laboratory of Distributed System Security Hubei Engineering Research Center on Big Data Security School of Cyber Science and Engineering Wuhan430074 China Huazhong University of Science and Technology 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 Wuhan430074 China

The Physical Unclonable Function (PUF) is valued for its lightweight nature and unique functionality, making it a common choice for securing hardware products requiring authentication and key generation mechanisms. In response to the susceptibility of individual PUFs to modeling attacks, advanced PUF variants have been developed to improve security measures. One notable type in this regard is the multiplexer-based composition of arbiter PUFs, known as MPUF, which aims to meet high reliability and security standards simultaneously. Current research on attacking MPUF encounters challenges such as substantial demands for training CRPs and low success rates. In this work, we propose a novel numerical modeling attack strategy for MPUFs. Using Kronecker products from mathematical perspective, this method precisely describe the MPUF model without using complex network architectures, boosting attack accuracy and overall efficiency. Experiment comparison with state-of-the-art works demonstrates that our method achieves better performance in terms of attack accuracy, robustness, and efficiency. Our method is able to successfully attack a (512, 8)-MPUF in 32.71 minutes with 97.14% accuracy, outperforming all existing attack methods on MPUFs. More, we validate our method through experiments with hardware implementations on FPGAs. The advantages of our method also include the adaptability to be employed to attack other MPUF variations like cMPUF and rMPUF, and the capability to be integrated with an existing attack method for launching efficient attack on MPUFs leveraging reliability information. © 2025 IEEE.

关键词： modeling attack multiplexer-based PUF Physical unclonable function

来源：评论

学校读者我要写书评

暂无评论

Dual Test-time Training for Out-of-distribution Recommender System

arXiv

引用

arXiv 2024年

作者： Yang, Xihong Wang, Yiqi Chen, Jin Fan, Wenqi Zhao, Xiangyu Zhu, En Liu, Xinwang Lian, Defu School of Computer National University of Defense Technology Changsha410073 China School of Business and Management Hong Kong University of Science and Technology Hong Kong The Hong Kong Polytechnic University Hong Kong Department of Data Science City University of Hong Kong Hong Kong Anhui Province Key Laboratory of Big Data Analysis and Application School of Computer Science and Technology University of Science and Technology of China Anhui Hefei230000 China

Deep learning has been widely applied in recommender systems, which has recently achieved revolutionary progress. However, most existing learning-based methods assume that the user and item distributions remain unchanged between the training phase and the test phase. However, the distribution of user and item features can naturally shift in real-world scenarios, potentially resulting in a substantial decrease in recommendation performance. This phenomenon can be formulated as an Out-Of-Distribution (OOD) recommendation problem. To address this challenge, we propose a novel Dual Test-Time-Training framework for OOD Recommendation, termed DT3OR. In DT3OR, we incorporate a model adaptation mechanism during the test-time phase to carefully update the recommendation model, allowing the model to adapt specially to the shifting user and item features. To be specific, we propose a self-distillation task and a contrastive task to assist the model learning both the user’s invariant interest preferences and the variant user/item characteristics during the test-time phase, thus facilitating a smooth adaptation to the shifting features. Furthermore, we provide theoretical analysis to support the rationale behind our dual test-time training framework. To the best of our knowledge, this paper is the first work to address OOD recommendation via a test-time-training strategy. We conduct experiments on five datasets with various backbones. Comprehensive experimental results have demonstrated the effectiveness of DT3OR compared to other state-of-the-art baselines. Copyright © 2024, 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

A DL Intrusion Detection Method for Low-Rate Denial of Service Attacks

A DL Intrusion Detection Method for Low-Rate Denial of Servi...

引用

Image Processing, Electronics and computers (IPEC), Asia-Pacific Conference on

作者： Yuhong Wu Yunshan Lv Xiangdong Hu Key Laboratory of Public Big Data Security Technology Chongqing School of Big Data & Computer Science Engineering Chongqing College of Mobile Communication China School of Computer Science and Technology Chongqing University of Posts and Telecommunications Chongqing China Key Laboratory of Public Big Data Security Technology School of Big Data & Computer Science Engineering Chongqing College of Mobile Communication Chongqing China School of Automation Chongqing University of Posts and Telecommunications Chongqing China

Denial of service (DoS) attacks is among the most harmful network threats. Low-rate DoS (LDoS) attacks feature good concealment and are likely to evade detection by conventional signal analysis-based methods. A LDoS attack detection method that combines the convolutional neural network and the gated recurrent unit is proposed in the present work. To evaluate the proposed method, we collect real user traffic from a university's website, label the traffic with LDoS attacks as normal traffic. In order to get the real attack traffic, this paper has collected some new LDoS attack programs, such as slowloris, slowhttptest, pwnlloris, torhammer and httplog, which can be obtained from public channels, and constructed a LAN environment that conformed to the working scenario of the program. By running the attack program, the attack traffic could be obtained. In a large sample scenario, the average detection rate of our proposed method can reach 98.68%. This article focuses on the research: the average detection rate of the detection method proposed in this article in the case that the security agency can only intercept a small number of attack samples. That is, in the small sample scenario, the detection of the collected traffic shows that the average detection rate of the proposed detection method can reach 89.54% in a small sample scenario. Its effective detection rate is obviously superior to the existing similar detection methods.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Three Preemption Approaches towards EDF Scheduling for Homogeneous Multiprocessors

Three Preemption Approaches towards EDF Scheduling for Homog...

引用

International Conference on Parallel and Distributed Systems (ICPADS)

作者： Peng Wu Zhi Li Zhicai Zhang Tao Yan Lu Chen Institute of Big Data Science and Industry Shanxi University Taiyuan China School of Computer and Information Technology Shanxi University Taiyuan China School of Computer Science and Technology Hainan University Haikou China

When scheduling a set of real-time tasks, researchers can choose between preemptive and non-preemptive algorithms. However, these algorithms each have their own advantages and drawbacks, necessitating specific analysis in different contexts. In contrast to these two extremes, researchers have proposed the concept of limited-preemption algorithms. Such limited-preemption algorithms serve as feasible alternatives to the former two approaches. Currently, research on this algorithm is based on a homogenous computing *** this paper, we investigate the fixed-preemption point method in limited-preemption algorithms. We compare the impact of limited-preemption EDF algorithm on the performance of homogenous multiprocessors under different fixed-preemption strategies. We introduce a preemptive strategy where high-priority tasks can only be preempted at the most suitable fixed-preemption points, and we provide an effective schedulability analysis for this strategy. The EDF scheduling algorithm based on the best-fit fixed preemption strategy is called the Best Preemptive EDF(BP-EDF). We have introduced an effective schedulability analysis for BP-EDF. Experimental results indicate that the strategy with the most suitable fixed-preemption points demonstrates superior schedulability compared to the other two strategies. Furthermore, its performance in terms of task preemption frequency closely approaches that of the strategy employing fixed-preemption at the lowest priority points.

关键词：

来源：评论

学校读者我要写书评

暂无评论

ML_BrainDetection: An Intelligent Model for Brain Tumor Identification Using Machine Learning 3

ML_BrainDetection: An Intelligent Model for Brain Tumor Iden...

引用

3rd International Mobile, Intelligent, and Ubiquitous Computing Conference, MIUCC 2023

作者： Abdelminaam, Diaa Salama Madbouly, Magda M. Farag, Mohamed S Gomaa, Ibrahim A Abd-Elghany Zeid, Magdy Abualigah, Laith Misr International University Faculty of Computer Science Cairo Egypt Benha University Faculty of Computers and Artificial Intelligence Benha Egypt Alexandria University Faculty of Computer and Data Science Alexandria Egypt Alobour High Institute for Computer and Informatics Computer Science Department Cairo Egypt Lebanese American University Department of Electrical and Computer Engineering Byblos13-5053 Lebanon Al-Ahliyya Amman University Hourani Center for Applied Scientific Research Amman19328 Jordan Middle East University Meu Research Unit Amman11831 Jordan School of Computer Sciences Universiti Sains Malaysia Pulau Pinang11800 Malaysia

ISBN: (纸本)9798350306231

Brain tumors are one of the greatest causes of death worldwide. Due to that, early diagnosis and classification of the tumor would surely help increase the chance of survival for patients worldwide. However, classifying and identifying brain tumors requires certain tools, including using Magnetic Resonance Imaging (MRI) to detect brain tumors. Moreover, many researchers in the past have also built up computer-Aided Diagnosis (CAD) systems to aid radiologists in their purpose of detecting brain abnormalities and tumors effectively. In this research, extracted multi-modal features like texture, morphological, entropy-based, Scale Invariant Features are proposed, and consequently, certain robust machine learning techniques, which are k-nearest neighbors (KNN), Decision Tree (DT), Support Vector Machine (SVM), Naive Bayes, as well as the Logistic Regression classifier were used to analyze the two data-sets included in the paper, detect tumors found in the brain, not to mention, examining their results, which have then shown the highest accuracy when using the Logistic regression classifier in both data sets, with the Support Vector Machine (SVM) classifier following it, which further support the argument made in the paper. © 2023 IEEE.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Multi-Grained Feature Pruning for Video-Based Human Pose Estimation

arXiv

引用

arXiv 2025年

作者： Wang, Zhigang Fan, Shaojing Liu, Zhenguang Wu, Zheqi Wu, Sifan Jiao, Yingying College of Computer Science and Technology Zhejiang Gongshang University Hangzhou China School of Computing National University of Singapore Singapore The State Key Laboratory of Blockchain and Data Security Zhejiang University Hangzhou China Institute of Blockchain and Data Security Hangzhou China College of Computer Science and Technology Jilin University Changchun China College of Computer Science and Technology Zhejiang University of Technology Hangzhou China

Human pose estimation, with its broad applications in action recognition and motion capture, has experienced significant advancements. However, current Transformer-based methods for video pose estimation often face challenges in managing redundant temporal information and achieving fine-grained perception because they only focus on processing low-resolution features. To address these challenges, we propose a novel multi-scale resolution framework that encodes spatio-temporal representations at varying granularities and executes fine-grained perception compensation. Furthermore, we employ a density peaks clustering method to dynamically identify and prioritize tokens that offer important semantic information. This strategy effectively prunes redundant feature tokens, especially those arising from multi-frame features, thereby optimizing computational efficiency without sacrificing semantic richness. Empirically, it sets new benchmarks for both performance and efficiency on three large-scale datasets. Our method achieves a 93.8% improvement in inference speed compared to the baseline, while also enhancing pose estimation accuracy, reaching 87.4 mAP on the PoseTrack2017 dataset. © 2025, CC BY.

关键词： Spatio-temporal data

来源：评论

学校读者我要写书评

暂无评论

Dpa-Ei: Long-Tailed Classification by Dual Progressive Augmentation from Explicit and Implicit Perspectives

SSRN

引用

SSRN 2024年

作者： Zhao, Yan Wenwei, He Zhao, Hong School of Computer Science Minnan Normal University Fujian Zhangzhou363000 China The Key Laboratory of Data Science and Intelligence Application Minnan Normal University Fujian Zhangzhou363000 China

Deep convolutional neural networks have significantly advanced visual recognition tasks;however, handling long-tailed distribution data remains a substantial challenge for machine learning. data augmentation is widely known as an effective technique for long-tailed image classification. However, most explicit and implicit augmentation methods have limitations that fail to balance sample-level generalization and feature-level semantic richness. In this paper, we present a dual progressive augmentation framework aimed at mitigating category imbalance in long-tailed classification scenarios. Our framework progressively enhances data diversity and representation by combining explicit and implicit augmentation stages. In the explicit augmentation stage, we adopt Mixup training to enhance sample-level generalization capacities, augmenting the diversity of tail-category data representations. Subsequently, the implicit augmentation stage incorporates category-level covariance learning and meta-learning methodologies, enriching feature-level semantic diversity. Extensive experimental results validate the effectiveness of our method, particularly for tail categories. The proposed framework boosts the performance of tail categories by 1%~2% over the second-best method while preserving the performance of head categories on the different long-tailed datasets. Code is available at http://***/fhqxa/DPA_EI. © 2024, 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

Snapshot boosting: a fast ensemble framework for deep neural networks

引用

science China(Information sciences) 2020年第1期63卷 77-88页

作者： Wentao ZHANG Jiawei JIANG Yingxia SHAO Bin CUI Center for Data Science Peking University National Engineering Laboratory for Big Data Analysis and Applications Department of Computer Science Beijing Key Lab of Intelligent Telecommunications Software and Multimedia School of Computer ScienceBeijing University of Posts and Telecommunications Key Lab of High Confidence Software Technologies Department of Computer SciencePeking University

Boosting has been proven to be effective in improving the generalization of machine learning models in many fields. It is capable of getting high-diversity base learners and getting an accurate ensemble model by combining a sufficient number of weak learners. However, it is rarely used in deep learning due to the high training budget of the neural network. Another method named snapshot ensemble can significantly reduce the training budget, but it is hard to balance the tradeoff between training costs and diversity. Inspired by the ideas of snapshot ensemble and boosting, we propose a method named snapshot boosting. A series of operations are performed to get many base models with high diversity and accuracy, such as the use of the validation set, the boosting-based training framework, and the effective ensemble strategy. Last, we evaluate our method on the computer vision(CV) and the natural language processing(NLP) tasks, and the results show that snapshot boosting can get a more balanced trade-off between training expenses and ensemble accuracy than other well-known ensemble methods.

关键词： ensemble learning deep learning boosting neural network snapshot ensemble

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：