检索结果-内蒙古大学图书馆

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Shibuya, Tatsukichi Inoue, Nakamasa Kawakami, Rei Sato, Ikuro Department of Computer Science Institute of Science Tokyo Tokyo Japan Department of Systems and Control Engineering Institute of Science Tokyo Tokyo Japan Denso IT Laboratory Tokyo Japan

ISBN: (纸本)9798350368741

For deploying deep neural networks on edge devices with limited resources, binary neural networks (BNNs) have attracted significant attention, due to their computational and memory efficiency. However, once a neural network is binarized, finetuning it on edge devices becomes challenging because most conventional training algorithms for BNNs are designed for use on centralized servers and require storing real-valued parameters during training. To address this limitation, this paper introduces binary stochastic flip optimization (BinSFO), a novel training algorithm for BNNs. BinSFO employs a parameter update rule based on Boolean operations, eliminating the need to store real-valued parameters and thereby reducing memory requirements and computational overhead. In experiments, we demonstrated the effectiveness and memory efficiency of BinSFO in finetuning scenarios on six image classification datasets. BinSFO performed comparably to conventional training algorithms with a 70.7% smaller memory requirement. Code is released at https://***/TatsukichiShibuya/ICASSP2025_BinSFO. © 2025 IEEE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Compression Format and Systolic Array Structure Co-design for Accelerating Sparse Matrix Multiplication in DNNs 24th

Compression Format and Systolic Array Structure Co-design f...

引用

24th International Conference on Algorithms and Architectures for Parallel Processing, ICA3PP 2024

作者： Cao, Yongxiang Jiang, Jixiang Zhao, Guocheng Wang, Wei Jiang, Hongxu Song, Yanfei School of Computer Science and Engineering Beihang University Beijing China Science and Technology on Complex System Control and Intelligent Agent Cooperation Laboratory Beijing China

ISBN: (纸本)9789819615445

The systolic array (SA) architecture is widely used in accelerator/AI chip design due to its excellent matrix multiplication acceleration effect. However, the mismatch between the traditional sparse matrix compression format and the SA architecture results in poor SA performance in computing sparse matrices. To solve the problem, we propose a method for co-designing the sparse compression format and SA architecture. This work proposes the Vector Group Compressed Coordinate (VGCC) data stream compression format and designs a sparse SA architecture based on hardware encoding and decoding. VGCC can handle data streams of any size without increasing the index bit width. The VGCC compression size is significantly smaller than general sparse compression formats such as COO/CSR. In a sparse matrix with a sparsity of 90%, it saves 67.13% of storage space compared to the state-of-the-art ECOO data stream compression format and achieves an average acceleration of 1.54 times. We also propose a VGCC-friendly SA architecture. Our SA architecture engine reduces logic consumption by 51.6% compared to Sparse TPU while achieving a 3.23x speedup. To mitigate the issue of high no-load rates in systolic arrays, we introduce a Data Transmission Priority (DTP) data matching algorithm. With sparsity levels ranging from 50% to 90%, the algorithm’s calculation speed is enhanced by 1.124x to 1.991x. We implemented a sparse DNN accelerator on an FPGA Zynq UltraScale+MPSoC ZCU102, utilizing the VGCC compression format proposed in this study for compressing input data. When compared to the S2 Engine, the accelerator achieved a 3.37x speed improvement in matrix multiplication with 75% sparsity, significantly enhancing the speed of sparse matrix multiplication. The accelerator achieved an average throughput of 1079.33GOP/s when inferencing DNNs. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Systolic arrays

来源：评论

学校读者我要写书评

暂无评论

Fully Distributed Model-Free Adaptive Sliding Mode control for MASs With Hybrid-Attacked Topology

引用

IEEE Transactions on Automation science and engineering 2025年 22卷 12336-12346页

作者： Duan, Shitao Chen, Guangdeng Zhou, Qi Li, Hongyi Huang, Tingwen Guangdong University of Technology School of Automation Guangdong-Hong Kong Joint Laboratory for Intelligent Decision and Cooperative Control Guangdong Provincial Key Laboratory for Intelligent Decision and Cooperative Control Guangzhou510006 China Southwest University College of Electronic and Information Engineering Chongqing400715 China Shenzhen University of Advanced Technology Faculty of Computer Science and Control Engineering Shenzhen518055 China

A fully distributed model-free adaptive sliding mode control (MFASMC) strategy is proposed in this paper for unknown nonlinear multi-agent systems (MASs), in which topology networks are exposed to hybrid attacks consisting of denial-of-service (DoS) and false data injection attacks. Hybrid attacks in topology networks can result in neighbor information dropouts and inaccuracies among agents. First, the MASs with unknown dynamics are translated to equivalent linear data equations by the dynamic linearization technique. Second, the impact of neighbor information dropouts caused by DoS attacks is mitigated by a designed attack compensation mechanism, in which the compensation error is guaranteed to be bounded in the sense of mathematical expectation. Then, a fully distributed MFASMC algorithm, which does not depend on knowledge of the Laplacian matrix, is designed to improve the robustness of MASs with topology networks exposed to hybrid attacks, thus indirectly mitigating the impact of neighbor information inaccuracies. Finally, the consensus error is rigorously proved to be bounded in the sense of mathematical expectation, and the validity of the proposed strategy is confirmed by simulations. © 2004-2012 IEEE.

关键词： Network topology

来源：评论

学校读者我要写书评

暂无评论

YOLOv8n-CGW: A novel approach to multi-oriented vehicle detection in intelligent transportation systems

引用

Multimedia Tools and Applications 2025年第7期84卷 3809-3840页

作者： Berwo, Michael Abebe Fang, Yong Sarwar, Nadeem Mahmood, Jabar Aljohani, Mansourah Elhosseini, Mostafa School of Information and Engineering Chang’an University Shaanxi Xi’an710064 China Department of Computer Science Bahria University Lahore Campus Punjab Lahore54600 Pakistan Department of Information Technology Faculty of Computing and Information Technology University of Sialkot Punjab Sialkot51040 Pakistan College of Computer Science and Engineering Taibah University Yanbu46421 Saudi Arabia Computers and Control Systems Engineering Department Faculty of Engineering Mansoura University Mansoura35516 Egypt

In the context of Intelligent Transportation Systems (ITS), the role of vehicle detection and classification is indispensable for streamlining transportation management, refining traffic control, and conducting in-depth accident analyses. However, the intricate task of accurately detecting multi-oriented vehicles in diverse scenarios remains a challenge, even with the advancements in ITS. Factors such as vehicle morphology, design, road conditions, structural details, and climatic variables provide a significant part in the detection. To address this, our study introduces an advanced YOLOv8n model, specifically the YOLOv8n-CGW, which is built upon pre-trained deep learning frameworks. This model is meticulously tailored to enhance the detection capabilities for multi-oriented vehicles. Comprehensive evaluations on esteemed datasets like MtV, MERGED, Custom dataset, DAWN, WEDGE, and UA-DETRAC underscore the superiority of our approach over existing models. Notably, our YOLOv8n-CGW model achieves a remarkable mAP of 83.3% and boasts a swift inference time of 0.0019s, setting a new benchmark in the domain. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

FineBID: Fine-Grained Protocol Reverse engineering for Bit-Level Field IDentification

引用

IEEE Transactions on Dependable and Secure Computing 2025年第3期22卷 2670-2686页

作者： Huang, Tao Gao, Yansong Zheng, Yifeng Wang, Zhanfeng Hu, Chao Fu, Anmin Nanjing University of Science and Technology School of Computer Science and Engineering Nanjing210094 China Xidian University State Key Laboratory of Integrated Services Networks Xian710071 China University of Western Australia CrawleyWA6009 Australia Harbin Institute of Technology School of Computer Science and Technology Guangdong Shenzhen518055 China Lexbell Company Ltd. Nanjing210014 China Army Engineering University of PLA College of Command and Control Engineering Nanjing210007 China

Protocol Reverse engineering (PRE) serves as the foundation for numerous security analysis techniques, such as vulnerability mining and intrusion detection, etc. The PRE analysis precision can directly affect the accuracy of these downstream techniques. The network-trace-based PRE technique has become the mainstream PRE technique attributed to its ease of implementation. However, without the prerequisite of additional dedicated devices or knowledge of information, the analysis precision of existing network-trace-based PRE methods is often achievable at only byte or half-byte level but not the fine-grained bit-level, which makes it increasingly challenging to meet the precision requirements of those downstream security applications. In this work, we propose a fine-grained PRE scheme, named FineBID, which makes the identification capability in a fine-grained manner for existing network-trace-based PRE methods into bit-level fields. FineBID follows the global characteristics of protocol fields and constructively models the bit-level field identification problem as a multi-objective decision model, which thus effectively overcomes the insufficient representativeness of bit-level fields’ local characteristics. Then, the multi-objective decision model is solved to obtain the Pareto solution set for different field segmentation levels, and the utility value per bit is further computed. The utility value can be used as the immediate indicator to determine whether each bit is a field boundary or not. Meanwhile, we propose an Actual Ground Truth that is more in line with the actual usage of each bit. With extensive experiments on the Internet, wireless, and industrial protocols, we affirm that FineBID can not only significantly reduce the search space for Ground Truth or Actual Ground Truth with a space reduction of 95.3% compared to exhaustive search, but also identify Ground Truth or Actual Ground Truth more accurately than other similar methods. © 2004-2012 IEEE.

关键词： Reverse engineering

来源：评论

学校读者我要写书评

暂无评论

Small Signal Stability-Guaranteed Moving Target Defense Against Load Redistribution Attack On IoT-based Smart Grid

引用

IEEE Internet of Things Journal 2025年第11期12卷 17413-17429页

作者： Wang, Bingdong Zhang, Zhenyong Wang, Mufeng Liu, Mengxiang Ye, Rui Deng, Ruilong Zhang, Xin Guizhou University State Key Laboratory of Public Big Data College of Computer Science and Technology Guiyang550025 China China Industrial Control Systems Cyber Emergency Response Team Beijing100040 China University of Sheffield Department of Automatic Control and System Engineering Sheffield United Kingdom Guizhou Power Grid Co. Ltd. Intelligent Operation Center China Zhejiang University State Key Laboratory of Industrial Control Technology College of Control Science and Engineering Hangzhou310027 China Huzhou Institute of Industrial Control Technology Huzhou313000 China

Moving target defense (MTD) is a promising approach to defend against load redistribution attacks on the internet-of-things (IoT)-based smart grid networks by probing the distorted state estimates with the distributed flexible AC transmission system. However, existing studies mainly focus on optimizing the performance of MTD and ignore the safety effect of it on the system’s operation. In this paper, we fill this gap by deeply analyzing the effect of MTD on the small signal stability and aim to alleviate the negative impact and guarantee its defending performance simultaneously. Firstly, the stability is formally described using the eigenvalue sensitivity. The relationship between the MTD-induced perturbation (MTDper) and the stability criteria is derived. Secondly, a new indicator is proposed to measure the effectiveness of MTDper. Thirdly, a constrained optimization problem is formulated to compute the bound of MTDper for guaranteeing the small signal stability. In addition, a surprising finding is that the stability margin can be improved and enhanced by optimizing the value of MTDper without losing the MTD’s effectiveness. Finally, we evaluate the performance of MTDper and its impact on the small signal stability with extensive simulations on the IEEE 30-bus, 39-bus, and 68-bus test power systems. © 2014 IEEE.

关键词： Constrained optimization

来源：评论

学校读者我要写书评

暂无评论

SKINC-NET: an efficient Lightweight Deep Learning Model for Multiclass skin lesion classification in dermoscopic images

引用

Multimedia Tools and Applications 2025年第13期84卷 12531-12557页

作者： Asif, Sohaib Qurrat-ul-Ain Khan, Saif Ur Rehman Amjad, Kamran Awais, Muhammad School of Computer Science and Engineering Central South University Hunan Changsha China School of Public Health Central South University Hunan Changsha China School of Control Science and Engineering Central South University Hunan Changsha China School of Minerals Processing and Bioengineering Central South University Hunan Changsha China

Diagnosing skin cancer through visual image examinations is time-consuming and error-prone, as the similar appearance and variations within each type of skin cancer challenge the efficiency and accuracy. computer-aided techniques, notably convolutional neural networks (CNNs), demand substantial computational resources and exhibit a high level of time complexity, rendering them impractical for real-world applications necessitating rapid and precise skin cancer detection. To address this issue, this research introduces SKINC-NET, a streamlined deep learning (DL) model specifically designed to achieve precise diagnoses of various skin cancer types while minimizing computational resource requirements. SKINC-NET can classify seven types of skin lesions using dermoscopic images, designed with dense layers, batch normalization (BN), and LeakyReLU activation functions. Numerous experiments were conducted on the HAM10000 dataset, using a data augmentation approach to tackle the class imbalance problem. The study compared the SKINC-NET model’s performance with four transfer learning models, namely ResNet50, VGG16, MobileNetV2, and EfficientNetB0. Several ablation studies were conducted to identify the optimal hyperparameters that lead to precise skin cancer diagnosis. The SKINC-NET model demonstrates its superiority by achieving the highest overall classification accuracy of 98.54% with minimal computation. The experimental findings affirm that the SKINC-NET model showcases remarkable robustness and presents several significant benefits, including lower computational complexity, a smaller number of trainable parameters, and superior performance compared to existing state-of-the-art models. SKINC-NET can assist dermatologists in making appropriate decisions regarding the early detection of skin cancer. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

A Zero-Sequence Current Analysis Approach for Rotating Machinery Fault Diagnosis of Induction Motor Drivetrain Based on Sparse Learning

引用

IEEE Transactions on Power Electronics 2025年第7期40卷 9800-9810页

作者： Liu, Zhao-Hua Long, Jun-Jie Wei, Hua-Liang Liu, Kan Zhang, Ying-Jie Li, Xiao-Hua Hunan University of Science and Technology School of Information and Electrical Engineering Xiangtan411201 China The University of Sheffield School of Electrical and Electronic Engineering Department of Automatic Control and Systems Engineering SheffieldS1 3JD United Kingdom Hunan University College of Mechanical and Vehicle Engineering Changsha410082 China Hunan University College of Computer Science and Electronic Engineering Changsha410082 China

Fault diagnosis of rotating machinery driven by induction motors has received increasing attention. Current diagnostic methods, which can be performed on existing inverters or current transformers of three-phase induction machines, have become a more economical and reliable alternative to vibration diagnostic methods. Existing references mainly focus on the stator current analysis of single-phase, but most single-phase current analysis methods utilize only a fraction of the total information accessible in the three-phase system. Field experience shows that zero-sequence current (ZSC) has more distinctive fault characteristics compared to single-phase current, and therefore it is better to use in mechanical fault detection and diagnostic tasks. This article proposes a novel sparse learning-based method for ZSC analysis for induction rotating motor drive fault diagnosis. First, it elaborates and compares the effectiveness of ZSC in revealing fault characteristics compared with single-phase current by analyzing the fault diagnosis mechanism of ZSC. In addition, the method proposes a correlation entropy-enhanced sparse learning method for the problem of high-frequency noise and interference in the ZSC signals, so as to enhance the learning of the features of the noise-containing signals. The diagnostic efficacy of the proposed method has been verified through experiments on two real datasets. © 1986-2012 IEEE.

关键词： Induction motors

来源：评论

学校读者我要写书评

暂无评论

A Framework for Real-Time Monitoring of Power Consumption of Distributed Calculation on Computational Cluster 15

A Framework for Real-Time Monitoring of Power Consumption of...

引用

15th International Conference on Cloud Computing and Services science, CLOSER 2025

作者： Krechowicz, Adam Faculty of Electrical Engineering Automatic Control and Computer Science Kielce University of Technology al. 1000-lecia PP 7 Kielce Poland

ISBN: (纸本)9789897587474

This paper proposes a framework for real-time monitoring of the power consumption of distributed calculation on the nodes of the cluster. The framework allows to visualize and analyze the provider results based on the context information about the performed calculation. The first part of the framework is devoted to monitoring the power consumption during the execution of machine learning algorithms and the performance of NoSQL storage. The second part is dedicated to the testing of distributed data storage, called Scalable Distributed Two-Layered Data Structure (SD2DS). The results show that the framework can be used in the development of a management system that could schedule computations to take full advantage of renewable energy. Copyright © 2025 by SCITEPRESS - science and Technology Publications, Lda.

关键词： Cluster computing

来源：评论

学校读者我要写书评

暂无评论

Consistent Counterfactual Explanations via Anomaly control and Data Coherence

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2025年第4期6卷 794-804页

作者： Movin, Maria Siciliano, Federico Ferreira, Rui Silvestri, Fabrizio Tolomei, Gabriele Spotify StockholmSE-111 Sweden Sapienza University of Rome Department of Computer Control and Management Engineering Rome00185 Italy Spotify LondonWC2N 6AT United Kingdom Sapienza University of Rome Department of Computer Science Rome00161 Italy

Algorithmic recourses are popular methods to provide individuals impacted by machine learning models with recommendations on feasible actions for a more favorable prediction. Most of the previous algorithmic recourse methods work under the assumption that the predictive model does not change over time. However, in reality, models in deployment may both be periodically retrained and have their architecture changed. Therefore, it is desirable that the recourse should remain valid when such a model update occurs, unless new evidence arises. We call this feature consistency. This article presents anomaly control and data coherence (ACDC), a novel model-agnostic recourse method that generates counterfactual explanations, i.e., instance-level recourses. ACDC is inspired by anomaly detection methods and uses a one-class classifier to aid the search for valid, consistent, and feasible counterfactual explanations. The one-class classifier asserts that the generated counterfactual explanations lie on the data manifold and are not outliers of the target class. We compare ACDC against several state-of-the-art recourse methods across four datasets. Our experiments show that ACDC outperforms baselines both in generating consistent counterfactual explanations, and in generating feasible and plausible counterfactual explanations, while still having proximity measures similar to the baseline methods targeting the data manifold. © 2024 IEEE.

关键词： Machine learning

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：