检索结果-内蒙古大学图书馆

SFedXL: Semi-Synchronous Federated Learning With Cross-Sharpness and Layer-Freezing

IEEE Internet of Things Journal 2025年第12期12卷 18692-18707页

作者： Zhao, Mingxiong Zhao, Shihao Feng, Chenyuan Yang, Howard H. Niyato, Dusit Quek, Tony Q. S. Yunnan University National Pilot School of Software Kunming650500 China Ministry of Education Engineering Research Center of Integration and Application of Digital Learning Technology Beijing100039 China Ministry of Education Engineering Research Center of Cyberspace Kunming650504 China Yunnan University of Finance and Economics Yunnan Key Laboratory of Service Computing Kunming650221 China EURECOM Sophia Antipolis06410 France Zhejiang University University of Illinois Urbana-Champaign Institute Zhejiang University Haining314400 China Nanyang Technological University School of Computer Science and Engineering 639798 Singapore Singapore University of Technology and Design Information Systems Technology and Design Pillar 487372 Singapore

Federated learning (FL) emerges as a potential solution for enabling multiple terminal devices to collaboratively accomplish computational tasks within an Unmanned Aerial Vehicle (UAV) swarm. However, traditional FL approaches, predicated on synchronous data aggregation, are not feasible for a UAV swarm owing to the inherently variable and dynamic nature of their communication networks compared with terrestrial systems. Furthermore, the data procured by UAVs is often highly heterogeneous, attributable to disparities in deployment environments and device attributes. Considering the distinct flight paths and unique operational conditions encountered by different UAVs, a considerable amount of data remains unlabeled. To tackle the challenges associated with asynchronous operations and the prevalence of unlabeled data, we introduce a novel framework termed Semi-synchronous FL with Cross-Sharpness and Layer-Freezing (SFedXL), tailored for a UAV swarm. In particular, we devise a cross-sharpness model training strategy aimed at optimizing the utilization of both labeled and unlabeled datasets. Additionally, we propose an innovative semi-synchronous model aggregation protocol, complemented by client-specific layer-freezing and client cluster scheduling, designed to expedite the training process. Our simulation results indicate that the proposed algorithm surpasses current FL methods in terms of object recognition accuracy and communication efficiency, albeit with a trade-off of increased local computation latency. © 2014 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Multimodal Fusion for Enhanced Parkinson's Disease Screening: Integrating Brain MRI, Hand Drawing, Facial Expressions, and Voice Analysis

Multimodal Fusion for Enhanced Parkinson's Disease Screening...

引用

Advancements in computing (ICAC), International Conference on

作者： Naveen Warnakulasuriya Shalinika De Silva Janani Madushika Sadali Gamage Sanath Jayawardena Anuradha Karunasena Department of Information Technology Faculty of Computing Sri Lanka Institute of Information Technology Malabe Sri Lanka Department of Computer Science and Software Engineering Faculty of Computing Sri Lanka Institute of Information Technology Malabe Sri Lanka Department of Computer Systems Engineering Faculty of Computing Sri Lanka Institute of Information Technology Malabe Sri Lanka

Parkinson's disease (PD) profoundly impacts millions in Sri Lanka, emphasizing the importance of early detection for better patient outcomes. We introduce “NeuraTrace PD,” an innovative application for early PD screening, combining brain MRI, hand-drawn image analysis, facial expression assessment, and voice analysis. For brain MRI analysis, we employed a Graph Convolutional Network (GCN), achieving an impressive accuracy of 91% and identifying distinct graph connectivity patterns using graph theory. We also use a deep Convolutional Neural Network (CNN) to reach 90% accuracy in analyzing hand-drawn images, while optimizing image resolution for efficiency and information retention. Facial expression analysis is a vital component, leveraging CNN and Constrained Local Models (CLM) to accurately extract Facial Action Units (FAUs), which contribute to a remarkable 95% accuracy using the XGBoost machine learning model. We introduce a Machine Learning (ML) algorithm and advanced signal processing to enhance voice feature analysis, offering a thorough understanding and classification of PD patients. It reaches 99% accuracy. In summary, our approach holds significant promise for advancing early PD diagnosis, ultimately improving the quality of life for those affected by this challenging condition.

关键词：

来源：评论

学校读者我要写书评

暂无评论

How often should one update control and estimation:review of networked triggering techniques

引用

Science China(Information Sciences) 2020年第5期63卷 6-23页

作者： Zhiyong CHEN Qing-Long HAN Yamin YAN Zheng-Guang WU School of Electrical Engineering and Computing University of Newcastle School of Software and Electrical Engineering Swinburne University of Technology Institute of Cyber-Systems and Control Zhejiang University

Management of resources is a constant topic in industrial systems. How to use minimum communication resources is of particular interest for control and estimation of networked systems. It raises the question for researchers in the field: how often should one update control and estimation? One of the most intelligent approaches is to trigger updates by events. In the literature, event-triggered control and estimation have been widely studied in the last decade. On one hand, events should be triggered sufficiently frequent to maintain system performance; on the other hand, the possibility of Zeno behavior caused by infinite frequency should be avoided. This review aims at revisiting some existing triggering techniques in a unified formulation, separated from system dynamics and control and estimation strategies. It brings readers better understanding of triggering mechanisms, the underlying technical challenges, and some promising future research topics.

关键词： event-triggered control event-triggered estimation networked systems Zeno behavior

来源：评论

学校读者我要写书评

暂无评论

Data Trading and Monetization: Challenges and Open Research Directions

arXiv

引用

arXiv 2024年

作者： Ramadan, Qusai Boukhers, Zeyd AlShaikh, Muath Lange, Christoph Jürjens, Jan Institute of Software Engineering University of Koblenz Germany Fraunhofer Institute for Applied Information Technology FIT Uniklinik Köln Germany College of Computing and Informatics Saudi Electronic University Saudi Arabia Fraunhofer Institute for Applied Information Technology FIT RWTH Aachen University Germany University of Koblenz Fraunhofer-Institute for Software and Systems Engineering ISST Germany

Traditional data monetization approaches face challenges related to data protection and logistics. In response, digital data marketplaces have emerged as intermediaries simplifying data transactions. Despite the growing establishment and acceptance of digital data marketplaces, significant challenges hinder efficient data trading. As a result, few companies can derive tangible value from their data, leading to missed opportunities in understanding customers, pricing decisions, and fraud prevention. In this paper, we explore both technical and organizational challenges affecting data monetization. Moreover, we identify areas in need of further research, aiming to expand the boundaries of current knowledge by emphasizing where research is currently limited or lacking. © 2024, CC BY.

关键词： Commerce

来源：评论

学校读者我要写书评

暂无评论

Formal Timing Analysis and Verification of the Interrupt-driven Programs for ISO26262

Formal Timing Analysis and Verification of the Interrupt-dri...

引用

2025 2nd International Conference on Algorithms, software Engineering and Network Security (ASENS)

作者： Qing She He Huang Shenglin Bao Xin Guo National Trusted Embedded Software Engineering Technology Research Center East China Normal University / Shanghai Huayuan Chuangxin Software Co. Ltd China High end Equipment Systems Department Shanghai Huayuan Chuangxin Software Co. Ltd China Basic Software Department East China Institute of Computing Technology China National Trusted Embedded Software Engineering Technology Research Center East China Normal University China

ISBN: (数字)9798331520298

ISBN: (纸本)9798331520304

Safety-critical systems refer to software systems whose operation can affect the safety of people and facilities used in the automotive industry, aerospace, and hospital health care systems. The high level of safety of software in the aforementioned fields has often drawn attention. ISO26262 has been issued for the automotive industry, which defines standards and procedures, including software level, software development standards, and software validation standards to ensure the high level of safety of software. In high-safety automotive software, a real-time safeguard requires the analysis of interrupt-driven programming. Such analysis presents a challenging and complex task because of the uncertainty of interrupts. In ISO26262, the formalized method is suggested to verify the process of software development. To achieve the verification goals defined in ISO26262, we use a formal method instead of traditional verification methods to analyze worst-case execution time. We analyze interrupt-driven programs from the perspective of timeliness and present the Interrupt Control Flow Graph (ICFG), which is used to describe the control flow of interrupt-driven programs. The execution of interrupts may largely increase the number of program states, potentially leading to state explosion, resulting in a sizable ICFG. To solve this problem, we propose an algorithm for reducing the size of ICFG. On the basis of the reduced ICFG, the worst-case execution path can be determined and thus be used to verify whether the program execution time exceeds the expected time and identify the execution path that may cause a timeout problem.

关键词： Industries Uncertainty software algorithms software systems Timing Flow graphs Standards Automotive engineering software development management software engineering

来源：评论

学校读者我要写书评

暂无评论

DA-Ada: Learning Domain-Aware Adapter for Domain Adaptive Object Detection

arXiv

引用

arXiv 2024年

作者： Li, Haochen Zhang, Rui Yao, Hantao Zhang, Xin Hao, Yifan Song, Xinkai Li, Xiaqing Zhao, Yongwei Li, Ling Chen, Yunji Intelligent Software Research Center Institute of Software CAS Beijing China State Key Lab of Processors Institute of Computing Technology CAS Beijing China State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation CAS Beijing China University of Chinese Academy of Sciences Beijing China

Domain adaptive object detection (DAOD) aims to generalize detectors trained on an annotated source domain to an unlabelled target domain. As the visual-language models (VLMs) can provide essential general knowledge on unseen images, freezing the visual encoder and inserting a domain-agnostic adapter can learn domain-invariant knowledge for DAOD. However, the domain-agnostic adapter is inevitably biased to the source domain. It discards some beneficial knowledge discriminative on the unlabelled domain, ***-specific knowledge of the target domain. To solve the issue, we propose a novel Domain-Aware Adapter (DA-Ada) tailored for the DAOD task. The key point is exploiting domain-specific knowledge between the essential general knowledge and domain-invariant knowledge. DA-Ada consists of the Domain-Invariant Adapter (DIA) for learning domain-invariant knowledge and the Domain-Specific Adapter (DSA) for injecting the domain-specific knowledge from the information discarded by the visual encoder. Comprehensive experiments over multiple DAOD tasks show that DA-Ada can efficiently infer a domain-aware visual encoder for boosting domain adaptive object detection. Our code is available at https://***/Therock90421/DA-Ada © 2024, CC BY-NC-SA.

关键词： Ada (programming language)

来源：评论

学校读者我要写书评

暂无评论

Fingerprint Agreement Using Enhanced Kerberos Authentication Protocol on M-Health

引用

Computer systems Science & Engineering 2022年第11期43卷 833-847页

作者： A.S.Anakath S.Ambika S.Rajakumar R.Kannadasan K.S.Sendhil Kumar School of Computing E.G.S.Pillay Engineering CollegeNagapattinam611002India Department of Computer Science and Engineering University College of Engineering AriyalurAriyalur621705India Department of Mathematics University College of Engineering AriyalurAriyalur621705India Department of Software Systems School of Computer Science and Engineering(SCOPE)Vellore Institute of Technology UniversityVellore632014India Department of IoT School of Computer Science and Engineering(SCOPE)Vellore Institute of Technology UniversityVellore632014India

Cloud computing becomes an important application development platform for processing user data with high *** providers are accustomed to providing storage centers outside the trusted location preferred by the data ***,ensuring the security and confidentiality of the data while processing in the centralized network is very *** secured key transmission between the sender and the receiver in the network is a huge challenge in managing most of the sensitive data transmission among the cloud *** are very active over the network like real authenticated user to hack the personal sensitive data,such as bank balance,health data,personal data,and confidential documents over the cloud *** this research,a secured key agreement between the sender and the receiver using Kerberos authentication protocol with fingerprint is proposed to ensure security in *** of patients are monitored using wireless sensor devices and are then transferred to the *** protocol helps in avoiding unnecessary communication of authenticated data over the cloud *** security process is a procedure with the best security in most of the authentication *** node is responsible in carrying data packets from the sender to the receiver in the cloud *** Kerberos protocol is used in trust node to ensure *** communication between the local health center and the healthcare server is ensured by using a fingerprint feature called minutiae form,which refers to the fingerprint image of both sender and *** computational and communicational cost of the proposed system is lesser when compared with other existing authentication methods.

关键词： Protocol security m-health cloud computing biometric fingerprint kerberos protocol

来源：评论

学校读者我要写书评

暂无评论

FAIR: Flow Type-Aware Pre-Training of Compiler Intermediate Representations

FAIR: Flow Type-Aware Pre-Training of Compiler Intermediate ...

引用

International Conference on software Engineering (ICSE)

作者： Changan Niu Chuanyi Li Vincent Ng David Lo Bin Luo State Key Laboratory for Novel Software Technology Nanjing University Nanjing China Human Language Technology Research Institute University of Texas at Dallas Richardson Texas USA School of Computing and Information Systems Singapore Management University Singapore

ISBN: (数字)9798400702174

ISBN: (纸本)9798350382143

While the majority of existing pre-trained models from code learn source code features such as code tokens and abstract syntax trees, there are some other works that focus on learning from compiler intermediate representations (IRs). Existing IR-based models typically utilize IR features such as instructions, control and data flow graphs (CDFGs), call graphs, etc. However, these methods confuse variable nodes and instruction nodes in a CDFG and fail to distinguish different types of flows, and the neural networks they use fail to capture long-distance dependencies and have over-smoothing and over-squashing problems. To address these weaknesses, we propose FAIR, a Flow type-Aware pre-trained model for IR that involves employing (1) a novel input representation of IR programs; (2) Graph Transformer to address over-smoothing, over-squashing and long-dependencies problems; and (3) five pre-training tasks that we specifically propose to enable FAIR to learn the semantics of IR tokens, flow type information, and the overall representation of IR. Experimental results show that FAIR can achieve state-of-the-art results on four code-related downstream tasks.

关键词： Codes Source coding Semantics Neural networks Syntactics Transformers Data models

来源：评论

学校读者我要写书评

暂无评论

On Evaluating the Efficiency of Source Code Generated by LLMs

On Evaluating the Efficiency of Source Code Generated by LLM...

引用

AI Foundation Models and software Engineering (Forge), IEEE/ACM International Conference on

作者： Changan Niu Ting Zhang Chuanyi Li Bin Luo Vincent Ng State Key Laboratory for Novel Software Technology Nanjing University Nanjing China School of Computing and Information Systems Singapore Management University Singapore Human Language Technology Research Institute University of Texas at Dallas Richardson Texas USA

ISBN: (数字)9798400705366

ISBN: (纸本)9798350388695

Recent years have seen the remarkable capabilities of large language models (LLMs) for code generation. Different from existing work that evaluate the correctness of the code generated by LLMs, we propose to further evaluate its efficiency. More efficient code can lead to higher performance and execution efficiency of pro-grams and software completed by LLM-assisted programming. First, we evaluate the efficiency of the code generated by LLMs on two benchmarks, HumanEval and MBPP. Then, we choose a set of pro-gramming problems from the online judge platform LeetCode to conduct a more difficult evaluation. Finally, we explore several prompts that would enable LLMs to generate more efficient code.

关键词： Codes Terminology Source coding Large language models Programming Benchmark testing software

来源：评论

学校读者我要写书评

暂无评论

Small sets of genuinely nonlocal Greenberger-Horne-Zeilinger states in multipartite systems

引用

Physical Review A 2024年第2期109卷 022428-022428页

作者： Zong-Xing Xiong Yongli Zhang Mao-Sheng Li Lvzhou Li Institute of Quantum Computing and Software School of Computer Science and Engineering Sun Yat-Sen University Guangzhou 510006 China School of Mathematics and Systems Science Guangdong Polytechnic Normal University Guangzhou 510665 China School of Mathematics South China University of Technology Guangzhou 510641 China

A set of orthogonal multipartite quantum states are called (distinguishability-based) genuinely nonlocal if they are locally indistinguishable across any bipartition of the subsystems. In this work, we consider the problem of constructing small genuinely nonlocal sets consisting of generalized Greenberger-Horne-Zeilinger (GHZ) states in multipartite systems. For system (C2)⊗N where N is large, using the language of group theory, we show that a tiny proportion Θ(1/2N) of the states among the N-qubit GHZ basis suffice to exhibit genuine nonlocality. Similar arguments also hold for the canonical generalized GHZ bases in systems (Cd)⊗N, wherever d is even and N is large. What is more, moving to the condition that any fixed N is given, we show that d+1 genuinely nonlocal generalized GHZ states exist in (Cd)⊗N, provided the local dimension d is sufficiently large. As an additional merit, within and beyond an asymptotic sense, the latter result also indicates some evident limitations of the “trivial othogonality-preserving local measurements” (TOPLM) technique that has been utilized frequently for detecting genuine nonlocality.

关键词： Nonlocality Quantum correlations in quantum information

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：