检索结果-内蒙古大学图书馆

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

作者： Xin Zou Zhengyu Zhou Jingyuan Xu Weiwei Liu School of Computer Science and National Engineering Research Center for Multimedia Software and Institute of Artificial Intelligence and Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University

ISBN: (纸本)9798331314385

ADABOOST is a well-known algorithm in boosting. Schapire and Singer propose, an extension of ADABOOST, named ***, for multi-class classification problems. Kégl shows empirically that *** works better when the classical one-against-all base classifiers are replaced by factorized base classifiers containing a binary classifier and a vote (or code) vector. However, the factorization makes it much more difficult to provide a convergence result for the factorized version of ***. Then, Kégl raises an open problem in COLT 2014 to look for a convergence result for the factorized ***. In this work, we resolve this open problem by presenting a convergence result for *** with factorized multi-class classifiers.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Dynamic Euclidean Bottleneck Matching

arXiv

引用

arXiv 2023年

作者： Abu-Affash, A. Karim Bhore, Sujoy Carmi, Paz Department of Software Engineering Shamoon College of Engineering Israel Department of Computer Science Indian Institute of Technology Bombay India Computer Science Department Ben-Gurion University Israel

A fundamental question in computational geometry is for a set of input points in the Euclidean space, that is subject to discrete changes (insertion/deletion of points at each time step), whether it is possible to maintain an approximate bottleneck matching in sublinear update time. In this work, we answer this question in the affirmative for points on a real line and for points in the plane with a bounded geometric spread. For a set P of n points on a line, we show that there exists a dynamic algorithm that maintains a bottleneck matching of P and supports insertion and deletion in O(log n) time. Moreover, we show that a modified version of this algorithm maintains a minimum-weight matching with O(log n) update (insertion and deletion) time. Next, for a set P of n points in the plane, we show that a (6√2)-factor approximate bottleneck matching of Pk, at each time step k, can be maintained in O(log ∆) amortized time per insertion and O(log ∆ + |Pk|) amortized time per deletion, where ∆ is the geometric spread of P. © 2023, CC BY.

关键词： Computational geometry

来源：评论

学校读者我要写书评

暂无评论

Digital Twin, Metaverse and Smart Cities in a Race to the Future

Digital Twin, Metaverse and Smart Cities in a Race to the Fu...

引用

Information Technology (ACIT)

作者： Hijab e Zainab Narmeen Zakaria Bawanay Department of Computer Science and Software Engineering Center for Computing Research Jinnah University for Women Karachi Pakistan

In the dynamic landscape of urban development, the synergy of the Metaverse, Digital Twins, and Smart Cities emerges as a potent catalyst for transformation. The Metaverse, a vibrant virtual space, collaborates with cutting-edge technologies, including Virtual Reality (VR), Augmented Reality (AR), the Internet of Things (IoT), and Artificial Intelligence (AI), crafting immersive experiences. Simultaneously, Digital Twins, serving as virtual replicas of physical entities, instigating an industrial revolution, while Smart Cities harness advanced technologies for sustainable urban living. This paper explores the applications of converging these technologies across diverse industries, showcasing how Digital Twins reshape conventional practices and amplify the concept of smart cities through the immersive realm of the Metaverse. However, the exploration unveils challenges, particularly in data privacy, interoperability, and ethical considerations. The study advocates for a balanced and informed approach, emphasizing that the symbiotic relationship between Digital Twins and the Metaverse should steer the evolution of Smart Cities toward security, innovation, and interconnected urban spaces.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Test-Time Backdoor Detection for Object Detection Models

arXiv

引用

arXiv 2025年

作者： Zhang, Hangtao Wang, Yichen Yan, Shihui Zhu, Chenyu Zhou, Ziqi Hou, Linshan Hu, Shengshan Li, Minghui Zhang, Yanjun Zhang, Leo Yu School of Cyber Science and Engineering Huazhong University of Science and Technology China School of Software Engineering Huazhong University of Science and Technology China School of Computer Science and Technology Huazhong University of Science and Technology China Harbin Institute of Technology China University of Technology Sydney Australia Griffith University Australia

Object detection models are vulnerable to backdoor attacks, where attackers poison a small subset of training samples by embedding a predefined trigger to manipulate prediction. Detecting poisoned samples (i.e., those containing triggers) at test time can prevent backdoor activation. However, unlike image classification tasks, the unique characteristics of object detection—particularly its output of numerous objects—pose fresh challenges for backdoor detection. The complex attack effects (e.g., "ghost" object emergence or "vanishing" object) further render current defenses fundamentally inadequate. To this end, we design TRAnsformation Consistency Evaluation (TRACE), a brand-new method for detecting poisoned samples at test time in object detection. Our journey begins with two intriguing observations: 1) poisoned samples exhibit significantly more consistent detection results than clean ones across varied backgrounds. 2) clean samples show higher detection consistency when introduced to different focal information. Based on these phenomena, TRACE applies foreground and background transformations to each test sample, then assesses transformation consistency by calculating the variance in objects confidences. TRACE achieves black-box, universal backdoor detection, with extensive experiments showing a 30% improvement in AUROC over state-of-the-art defenses and resistance to adaptive attacks. Copyright © 2025, The Authors. All rights reserved.

关键词： Image classification

来源：评论

学校读者我要写书评

暂无评论

Multi-Objective Optimization Design for DC-Link Capacitor of Motor Driver

Multi-Objective Optimization Design for DC-Link Capacitor of...

引用

2024 CPSS and IEEE International Symposium on Energy Storage and Conversion, CPSS and ISESC 2024

作者： Liu, Qian Huang, Kangxing Wei, Xueliang Zheng, Yihong Xiang, Hao School of Automation Wuhan University of Technology Wuhan China Hong Kong University of Science and Technology Department of Electronic and Computer Engineering Hong Kong Hong Kong College of Artificial Intelligence China University of Petroleum-Beijing Beijing China Wuhan IActuator Robotics Co. Ltd Software Development Department Wuhan China

ISBN: (纸本)9798350380514

The capacitor is a common weak point in motor drive system design, necessitating targeted selection, design, and optimization of capacitor banks within these systems. Traditional capacitor design often relies on engineering experience, usually meeting specific capacitance and voltage requirements through simple parallel or series connections of capacitors. This approach often fails to fully optimize volume, cost, and reliability. This paper uses comprehensive theoretical modeling to optimize capacitor module design, effectively balancing capacitor lifespan, cost, and volume. For the design and selection of DC-link capacitors in a 1kW permanent magnet synchronous motor drive system, a multi-objective optimization design method for capacitor banks is proposed. The study treats the individual capacitance values of parallel capacitors as variables, significantly enhancing the overall performance of the capacitor module through precise optimization and the introduction of necessary constraints in the selection design. The proposed method is validated with a 1kW permanent magnet synchronous motor drive system. The optimization results show that under the guidance of a multi-objective optimization framework, the designed DC-link capacitor module meets physical size requirements while achieving overall optimization in terms of volume, cost, and lifespan, significantly improving system reliability. © 2024 IEEE.

关键词： Permanent magnets

来源：评论

学校读者我要写书评

暂无评论

Verilua: An Open Source Versatile Framework for Efficient Hardware Verification and Analysis Using LuaJIT

Verilua: An Open Source Versatile Framework for Efficient Ha...

引用

2025 Design, Automation and Test in Europe Conference, DATE 2025

作者： Cai, Ye Zheng, Chuyu He, Wei Tang, Dan College of Computer Science and Software Engineering Shenzhen University Shenzhen China Beijing Institute of Open Source Chip Beijing China Beijing China

ISBN: (纸本)9783982674100

The growing complexity of hardware verification highlights limitations in existing frameworks, particularly regarding flexibility and reusability. Current methodologies often require multiple specialized environments for functional verification, waveform analysis, and simulation, leading to toolchain fragmentation and inefficient code reuse. This paper presents Verilua, a unified framework leveraging LuaJIT and the Verilog Procedural Interface (VPI), which integrates three core functional-ities: Lua-based functional verification, a scripting engine for RTL simulation, and waveform analysis. By enabling complete code reuse through a unified Lua codebase, the framework achieves a 12x speedup in RTL simulation compared to cocotb and a 70x improvement in waveform analysis over state-of-the-art solutions. Through consolidating verification tasks into a single platform, Verilua enhances efficiency while reducing tool fragmentation and learning overhead, addressing critical challenges in modern hardware design. © 2025 EDAA.

关键词： computer software reusability

来源：评论

学校读者我要写书评

暂无评论

Medical Image Analysis of Lung Tumor Diagnosis Based on Generalized Fuzzy C-Means Clustering Algorithm

Medical Image Analysis of Lung Tumor Diagnosis Based on Gene...

引用

International Conference on Emerging Electronic and Automation Technology 2022, ICEEAT 2022

作者： Zwaid, Jassim Mohammed Ahmed Murad, Masrah Azrifah Azmi Khalid, Fatimah Binti Manshor, Noridayu Al-Jumaily, Abdulmajeed Faculty of Computer Science and Information Technology University Putra Malaysia UPM Serdang 43400 Malaysia Affiliated to University of Anbar Ramadi Iraq Department of Software Engineering and Information System Faculty of Computer Science and Information Technology University Putra Malaysia UPM Serdang 43400 Malaysia Multimedia Department Faculty of Computer Science and Information Technology University Putra Malaysia UPM Serdang 43400 Malaysia Department of Signal Theory and Communications Universidad Carlos III de Madrid Leganés Madrid28911 Spain

In recent years, the rapid development of medical imaging technology has brought medical image analysis into the era of big data. CT imaging technology is one of the most common imaging methods for disease screening. This paper introduces medical image analysis of lung tumor diagnosis based on Generalized fuzzy c - means clustering algorithm was proposed. Brief introduce lung CT imaging to diagnose lung tumors requires a huge workload, which is often accompanied by long-term image reading and subjective evaluation. The experimental data comes from 15 patient samples which include two groups of samples are selected as test objects. The result shows the positive rate is high, resulting in misdiagnosis. In addition, great success of deep learning technology in the field of computer vision has made it possible to realize computer-aided diagnosis and screening of lung cancer. © Published under licence by IOP Publishing Ltd.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Application of Neural Networks for Network Traffic Monitoring and Analysis 8

Application of Neural Networks for Network Traffic Monitorin...

引用

8th International Symposium on Multidisciplinary Studies and Innovative Technologies, ISMSIT 2024

作者： Kushnerov, Oleksandr Murr, Pierre Herasymov, Serhii Milevskyi, Stanislav Melnyk, Marharyta Golovashych, Sergii Sumy State University Department of Economic Cybernetics Sumy Ukraine International University of Science and Technology in Kuwait Computer Engineering Department Kuwait Kuwait 'Kharkiv polytechnic institute' Cybersecurity department National Technical University Kharkiv Ukraine Set University Department of Cyber Security and information protection Kyiv Ukraine National Technical University 'Kharkiv polytechnic institute' Department of Software Engineering and Management Intelligent Technologies Kharkiv Ukraine

ISBN: (纸本)9798350354423

Monitoring and analyzing network traffic are critically important tasks in the context of increasing cyber threats. Traditional data processing methods often struggle to handle the large volumes and complexity of modern network traffic, necessitating the use of advanced technologies such as neural networks and machine learning. This article demonstrates the application of neural networks for monitoring and analyzing network traffic, highlighting their ability to learn from large datasets to detect complex patterns and anomalies. Using TensorFlow, we developed a neural network model based on an autoencoder that effectively identifies and classifies anomalies in network traffic. The model's effectiveness was validated using a synthetic dataset from Kaggle, showing high accuracy on both training and testing data. Our approach significantly improves the efficiency and accuracy of threat detection compared to traditional methods, making it a valuable tool for enhancing cybersecurity. © 2024 IEEE.

关键词： Cyber attacks

来源：评论

学校读者我要写书评

暂无评论

Blockchain-Enabled EHR Framework for Internet of Medical Things

引用

computers, Materials & Continua 2021年第4期67卷 211-221页

作者： Lewis Nkenyereye S.M.Riazul Islam Mahmud Hossain M.Abdullah-Al-Wadud Atif Alamri Department of Computer and Information Security Sejong UniversitySeoul05006Korea Department of Computer Science and Engineering Sejong UniversitySeoul05006Korea Department of Computer Science University of Alabama at BirminghamAlabama35294USA Department of Software Engineering King Saud UniversityRiyadh11543Saudi Arabia

The Internet of Medical Things(IoMT)offers an infrastructure made of smart medical equipment and software applications for healthcare *** the internet,the IoMT is capable of providing remote medical diagnosis and timely health *** patients can use their smart devices to create,store and share their electronic health records(EHR)with a variety of medical personnel including medical doctors and ***,unless the underlying commination within IoMT is secured,malicious users can intercept,modify and even delete the sensitive EHR data of *** also lose full control of their EHR since most healthcare services within IoMT are constructed under a centralized platform outsourced in the ***,it is appealing to design a decentralized,auditable and secure EHR system that guarantees absolute access control for the patients while ensuring privacy and *** the features of blockchain including decentralization,auditability and immutability,we propose a secure EHR framework which is mainly maintained by the medical *** this framework,the patients’EHR data are encrypted and stored in the servers of medical institutions while the corresponding hash values are kept on the *** make use of security primitives to offer authentication,integrity and confidentiality of EHR data while access control and immutability is guaranteed by the blockchain *** security analysis and performance evaluation of the proposed framework confirms its efficiency.

关键词： Internet of Medical Things security privacy preservation blockchain access control

来源：评论

学校读者我要写书评

暂无评论

Research on CNN-Based Malicious Traffic Identification Method 2021

Research on CNN-Based Malicious Traffic Identification Metho...

引用

7th International Conference on Computing and Artificial Intelligence, ICCAI 2021

作者： Tang, BinHui Jiang, Wen Xu Ke, Yong Xiang Jincheng College of Sichuan University School of Computer and Software Sichuan University School of Cyber Science and Engineering China Jincheng College of Sichuan University School of Computer and Software China

ISBN: (纸本)9781450389501

Today's Internet environment is full of all kinds of normal and malicious traffic, how to identify the feature categories of malicious traffic plays a crucial role in network management and security. With the rapid growth of modern Internet traffic, classical machine learning methods are limited by efficiency and functionality and are no longer sufficient to deal with large and complex network traffic. Thus, we propose a convolutional neural network-based traffic identification method. Firstly, the full traffic dataset for testing is preprocessed with one-hot coding. Secondly, a convolutional neural network model is built for testing, a softmax classifier is used to detect and classify the normal traffic and various malicious traffic, and finally, obtain the classification results of each traffic type. The experimental tests on the publicly available dataset show that the detection accuracy of malicious traffic is close to 99%, and the loss value is less than 1%. © 2021 ACM.

关键词： Convolutional neural networks

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：