检索结果-内蒙古大学图书馆

Overcoming Forgetting Using Adaptive Federated Learning for IIoT Devices With Non-IID Data

IEEE Internet of Things Journal 2025年

作者： Zhang, Benteng Mao, Yingchi Xu, Haowen Chen, Yihan Muazu, Tasiu He, Xiaoming Wu, Jie Hohai University College of Computer Science and Software Engineering Nanjing211100 China Nanjing University of Posts and Telecommunications College of Internet of Things Nanjing210003 China Temple University Center for Networked Computing PhiladelphiaPA19122 United States

In real-world Industrial Internet of Things (IIoT) scenarios, due to the limited storage capacity of IIoT devices, fresh data continuously received by diverse devices will overwrite the outdated data and change the local data distribution. However, state-of-the-art studies have demonstrated that Federated Learning (FL) tends to focus on training with fresh data, and the latest global model may forget the historical update directions (i.e., catastrophic forgetting). This issue can significantly degrade the global model accuracy. Existing methods primarily focus on integrating outdated data characteristics into fresh data but overlook the large parameter update gap between global and local models during global aggregation. This gap can cause the global model updates to deviate from the optimal direction. To this end, we propose a federated adaptive weighted aggregation method based on model consistency (FedAWAC). Specifically, FedAWAC measures the model consistency on devices and dynamically adjusts the aggregation weights of each local model, thereby guiding the global model toward optimal updates. Furthermore, FedAWAC integrates M historical global models most correlated to the latest global model on the cloud server to overcome catastrophic forgetting. Experiments on 4 different datasets (Non-IID settings) indicate that compared to 5 baselines, FedAWAC can improve global model accuracy by an average of 1.86%, reduce the forgetting rate by an average of 3.91%, and save average memory usage by up to 2.57GB. © 2025 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Securing the Silver Surfers: Cybersecurity Awareness Among Older Adults

Securing the Silver Surfers: Cybersecurity Awareness Among O...

引用

International Conference on Ubiquitous Information Management and Communication (IMCOM)

作者： Haslinda Sutan Ahmad Nawi Siti Fatimah Omar Suzana Basaruddin Info. Science & Computing Department Management and Science University Selangor Malaysia Software Engineering Section UniKL MIIT Kuala Lumpur Malaysia Informatics & Analytics Section UniKL MIIT Kuala Lumpur Malaysia

ISBN: (数字)9798331507817

ISBN: (纸本)9798331507824

This research investigates the perceptions of cybersecurity among older adults in Malaysia, aged 60 years and above, who are reliant on technology for basic activities. The data from 331 participants showed that 30.8% were online for 4–8 hours daily, while 3.3% were completely disconnected. Interestingly, more than half of the respondents were unaware of cybersecurity threats, especially malware and ransomware, which put their data in jeopardy. While there is some awareness related to unauthorized access and computer fraud, knowledge gaps remain significant, and effective educational interventions are thus needed. Informal, self-directed learning appears to be a widespread way of acquiring skills in this age group, often supplemented by community workshops and peer networks. Given that almost half of the older adults spend considerable time online, improving their digital skills in tandem with their support systems is an important goal for enhancing their online experiences. It indicates the need for a twin approach: enhancing cybersecurity awareness with specialized educational programs, alongside developing flexible opportunities for informal learning. Educational institutions and relevant policies are encouraged to prioritize building hands-on, specific cybersecurity education for seniors. Also, user interface design considering cybersecurity-related issues can make using the technological environment as user-friendly and secure as possible. These strategies will contribute to creating an inclusive digital society, where older adults are confident and safe in using digital technologies while continuously contributing to their personal and professional development.

关键词： Resistance Educational programs Silver Conferences User interfaces Fraud Peer-to-peer computing Ransomware Computer security Older adults

来源：评论

学校读者我要写书评

暂无评论

Learnable Prompt as Pseudo-Imputation: Rethinking the Necessity of Traditional EHR Data Imputation in Downstream Clinical Prediction 25

Learnable Prompt as Pseudo-Imputation: Rethinking the Necess...

引用

Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.1

作者： Weibin Liao Yinghao Zhu Zhongji Zhang Yuhang Wang Zixiang Wang Xu Chu Yasha Wang Liantao Ma School of Computer Science Peking University Beijing China & Key Laboratory of High Confidence Software Technologies Ministry of Education Beijing China National Engineering Research Center for Software Engineering Peking University Beijing China School of Software and Microelectronics Peking University Beijing China & Key Laboratory of High Confidence Software Technologies Ministry of Education Beijing China School of Computer Science Peking University Beijing China Center on Frontiers of Computing Studies Peking University Beijing China Key Laboratory of High Confidence Software Technologies Ministry of Education Beijing China & Peking University Information Technology Institute (Tianjin Binhai) Peking University Tianjin China National Engineering Research Center for Software Engineering Peking University Beijing China & Key Laboratory of High Confidence Software Technologies Ministry of Education Beijing China

ISBN: (纸本)9798400712456

Analyzing the health status of patients based on Electronic Health Records (EHR) is a fundamental research problem in medical informatics. The presence of extensive missing values in EHR makes it challenging for deep neural networks (DNNs) to directly model the patient's health status. Existing DNNs training protocols, including Impute-then-Regress Procedure and Jointly Optimizing of Impute-n-Regress Procedure, require the additional imputation models to reconstruction missing values. However, Impute-then-Regress Procedure introduces the risk of injecting imputed, non-real data into downstream clinical prediction tasks, resulting in power loss, biased estimation, and poorly performing models, while Jointly Optimizing of Impute-n-Regress Procedure is also difficult to generalize due to the complex optimization space and demanding data requirements. Inspired by the recent advanced literature of learnable prompt in the fields of NLP and CV, in this work, we rethought the necessity of the imputation model in downstream clinical tasks, and proposed Learnable Prompt as Pseudo-Imputation (PAI) as a new training protocol to assist EHR analysis. PAI no longer introduces any imputed data but constructs a learnable prompt to model the implicit preferences of the downstream model for missing values, resulting in a significant performance improvement for all state-of-the-arts EHR analysis models on four real-world datasets across two clinical prediction tasks. Further experimental analysis indicates that PAI exhibits higher robustness in situations of data insufficiency and high missing rates. More importantly, as a plug-and-play protocol, PAI can be easily integrated into any existing or even imperceptible future EHR analysis models. The code of this work is deployed publicly available at https://***/MrBlankness/PAI to help the research community reproduce the results and assist the EHR analysis tasks.

关键词： electronic health records

来源：评论

学校读者我要写书评

暂无评论

Rsta: A Reversible Signal Transmission Architecture for Breaking the Limitation of Unidirectional Transmission

SSRN

引用

SSRN 2025年

作者： Liu, Xin Cui, Shuang Zhang, Xun Zhou, Fuxing Sheng, Kaikai Wang, Bin Zhang, Qiang School of Computer Science and Technology Dalian University of Technology Dalian116024 China Key Laboratory of Advanced Design and Intelligent Computing Ministry of Education School of Software Engineering Dalian University Dalian116622 China

Molecular circuits are essential tools for biosensing and molecular information processing. However, the long-distance and reversible transmission of orthogonal signals remains a significant challenge in molecular circuits, preventing molecular signals from achieving reversible transmission similar to that of electrical signals in wires. To overcome the limitations of unidirectional signal transmission and facilitate the integration of antagonistic circuits. We propose a reversible signal transmission architecture (RSTA) with wire-like characteristics to achieve reversible signal transmission based on fundamental molecular devices known as reversible signal fan-out devices (RSFDs). The polarity of the input signal dictates the transmission direction, establishing the initial condition for the output signals. Cascaded RSFDs within RSTA offer two key advantages. First, the temporal response to signals. Second, the reversible transformation of opposite circuit types without structural modification. We experimentally validated the principles and functionality of RSTA using temporal logic device and integrated molecular circuits. This work advances the field of reversible molecular signal transmission, enabling the rational design of complex molecular circuits and molecular information processing systems. We envision applications of this technology in nanomachines, molecular computing, and biosensing. © 2025, The Authors. All rights reserved.

关键词： Temporal logic

来源：评论

学校读者我要写书评

暂无评论

Zram Instance Pool Framework for Adaptive Memory Compression in Resource-Sensitive Embedded Operating Systems

引用

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

作者： Deng, Yin Xie, Guoqi Xiong, Chenglai Zhao, Sirong Ren, Wei Li, Kenli Hunan University Key Laboratory for Embedded and Network Computing of Hunan Province College of Computer Science and Electronic Engineering Hunan Changsha410082 China Huawei Technology Co. Ltd. Central Software Institute Guangdong Shenzhen518129 China

Memory compression can reduce the size of the inactive data in the random access memory (RAM), thereby freeing up unused space and allowing more programs to run;however, current mainstream memory compression frameworks (e.g., Zram and Zswap) and algorithms (e.g., Zstd and Lz4) do not effectively solve the problem of increased CPU utilization, causing they cannot be directly applied to the resource-sensitive embedded operating system that is sensitive to both CPU utilization and memory usage. In this study, we develop a Zram instance pool framework called ZramPool for adaptive memory compression. The framework consists of the swap space with multiple Zram instances and the adaptive Zram compression module. Through introducing linear regression analysis, the number of Zram instances can be adaptively adjusted based on the size of the compressed data, allowing Zram instances to work in parallel to match the workload. In ZramPool, we achieve two different requirements of reducing CPU utilization while keeping compression speed and increasing compression speed while keeping CPU utilization. ZramPool is deployed in the embedded Linux OS with a 8GB memory size running on the ARMv8 architecture. For the first requirement, ZramPool can reduce CPU utilization by an average of 11.42% while the compression speed only decreases by an average of 2.4%. For the second requirement, ZramPool can increase compression speed by an average of 11.71% while the CPU utilization only increases by an average of 1.9%. © 1982-2012 IEEE.

关键词： Random access storage

来源：评论

学校读者我要写书评

暂无评论

The role of moral intention and moral obligation in predicting attitudes toward avoiding AIgiarism: A protection motivation theory perspective

引用

Computers in Human Behavior Reports 2025年 18卷

作者： Ali, Ismail Mohamed Arpaci, Ibrahim Faculty of Computing SIMAD University Somalia Department of Software Engineering Faculty of Engineering and Natural Sciences Bandirma Onyedi Eylul University Balikesir Turkey Department of Computer Science and Engineering College of Informatics Korea University Seoul 02841 South Korea

Traditional methods of combating plagiarism have proven ineffective due to the dual use of AI chatbots to plagiarize and avoid detection. Therefore, there is a growing rationale to focus on moral responsibility towards protecting academic integrity. This study investigated the impact of students' moral intentions and moral obligations on their attitudes toward avoiding AI-assisted plagiarism (AIgiarism). Data was collected from a representative group of 263 students and analyzed using “partial least squares structural equation modeling” (PLS-SEM) and “artificial neural network” (ANN). The results revealed that students' moral intentions and obligations positively influenced their attitudes towards avoiding AIgarism. “Protection motivation theory” (PMT) constructs positively predicted students' moral intentions. Moreover, the results of the ANN analysis showed that moral obligation and intention were the most critical factors influencing students’ attitudes toward avoiding plagiarism. This study will help educational institutions develop AI-supported anti-plagiarism solutions in areas such as identification, instilling moral responsibility, and creating an environment of trust and support. © 2025 The Authors

关键词： AIgiarism Moral intention Moral obligation Plagiarism Protection motivation theory

来源：评论

学校读者我要写书评

暂无评论

Preserving Location Privacy of IoT Devices in Heterogeneous Edge computing Architecture Through Deniability-Based Authentication

引用

IEEE Transactions on Consumer Electronics 2025年

作者： Ali, Ikram Li, Jianqiang Chen, Jie Chen, Yong Ullah, Shamsher Wakeel, Abdul Shenzhen University National Engineering Laboratory of Big Data System Computing Technology Shenzhen518060 China Shenzhen University College of Computer Science and Software Engineering Shenzhen518060 China University of Electronic Science and Technology of China School of Automation Engineering Chengdu611731 China National University of Sciences and Technology Military College of Signals Department of Electrical Engineering Islamabad Pakistan

Edge computing moves cloud services closer to consumer Internet of Things (IoT) devices, reducing latency and bandwidth usage. This setup enables faster responses but also introduces new security challenges, particularly concerning location privacy when sender and receiver use different security techniques. The authentication process in such architectures can expose the location of IoT devices. To address this issue, researchers have proposed various privacy-preserving schemes. However, these are computationally heavy for IoT devices. To tackle this issue, we propose two novel heterogeneous deniable authentication schemes: HDA-IoT-I and HDA-IoT-II. These schemes use public-key infrastructure and identity-based cryptography to protect the location privacy of IoT devices in heterogeneous edge computing environments. Our schemes enable the designated edge device to verify the source of a message without being able to prove its origin to a third party, thus preserving privacy. Additionally, they incorporate a batch verification method, which speeds up the verification of multiple deniable authenticators. The proposed schemes are formally proven secure in the random oracle model based on the hardness assumption of elliptic curve inverse computational Diffie-Hellman problem. The performance evaluation demonstrate that our schemes significantly enhance computational and communication efficiency, making them suitable for IoT devices. © 2025 IEEE.

关键词： Differential privacy

来源：评论

学校读者我要写书评

暂无评论

DHD: Double Hard Decision Decoding Scheme for NAND Flash Memory

DHD: Double Hard Decision Decoding Scheme for NAND Flash Mem...

引用

Design, Automation and Test in Europe Conference and Exhibition

作者： Lanlan Cui Yichuan Wang Renzhi Xiao Miao Li Xiaoxue Liu Xinhong Hei School of Computer Science and Engineering XI'AN University of Technology Xi'an China Shaanxi Key Laboratory for Network Computing and Security Technology Xi'an China School of Software Engineering Jiangxi University of Science and Technology Nanchang China School of Information Engineering Zhongnan University of Economics and Law Wuhan China

ISBN: (数字)9783982674100

ISBN: (纸本)9798331534646

With the advancement of NAND flash technology, the increased storage density leads to intensified interference, which in turn raises the error rate during data retrieval. To ensure data reliability, low-density parity-check (LDPC) codes are extensively employed for error correction in NAND flash memory. Although LDPC soft decision decoding offers high error correction capability, it comes with a significant latency. Conversely, hard-decision decoding, although faster, lacks sufficient error correction strength. Consequently, flash memory typically initiates with hard-decision decoding and resorts to multiple soft decision decoding upon failure. To minimize decoding latency, this paper proposes a decoding mechanism based on the double hard decision, called DHD. This DHD scheme improves the Log-Likelihood Ratio (LLR) in the hard decision process. After the first hard decision fails, the read reference voltage (RRV) is adjusted to perform the second hard decision decoding. If the second hard decision also fails, soft decision decoding is then employed. Experimental results demonstrate that when the Raw Bit Error Rate (RBER) is $8.5 \times 10^{-3}$ , DHD reduces the Frame Error Rate (FER) by 86.4% compared to the traditional method.

关键词： Lead acid batteries Europe Voltage Interference Decoding Error correction Error correction codes Reliability Flash memories Iterative decoding

来源：评论

学校读者我要写书评

暂无评论

DEGSTalk: Decomposed Per-Embedding Gaussian Fields for Hair-Preserving Talking Face Synthesis

DEGSTalk: Decomposed Per-Embedding Gaussian Fields for Hair-...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Kaijun Deng Dezhi Zheng Jindong Xie Jinbao Wang Weicheng Xie Linlin Shen Siyang Song Computer Vision Institute School of Computer Science and Software Engineering Shenzhen University National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Guangdong Provincial Key Laboratory of Intelligent Information Processing Department of Computer Science University of Exeter

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Accurately synthesizing talking face videos and capturing fine facial features for individuals with long hair presents a significant challenge. To tackle these challenges in existing methods, we propose a decomposed per-embedding Gaussian fields (DEGSTalk), a 3D Gaussian Splatting (3DGS)-based talking face synthesis method for generating realistic talking faces with long hairs. Our DEGSTalk employs Deformable Pre-Embedding Gaussian Fields, which dynamically adjust pre-embedding Gaussian primitives using implicit expression coefficients. This enables precise capture of dynamic facial regions and subtle expressions. Additionally, we propose a Dynamic Hair-Preserving Portrait Rendering technique to enhance the realism of long hair motions in the synthesized videos. Results show that DEGSTalk achieves improved realism and synthesis quality compared to existing approaches, particularly in handling complex facial dynamics and hair preservation. Our code is available at https://***/CVI-SZU/DEGSTalk.

关键词： Hair Training Three-dimensional displays Dynamics Signal processing Rendering (computer graphics) Noise measurement Speech processing Faces Videos

来源：评论

学校读者我要写书评

暂无评论

Pattern recognition in programming to support the evaluation of Computational Thinking

Pattern recognition in programming to support the evaluation...

引用

IEEE World engineering Education Conference (EDUNINE)

作者： Steven Pacheco-Portuguez Antonio Gonzalez-Torres Lilliana Sancho-Chavarria Ignacio Trejos-Zelaya Jorge Monge-Fallas School of Computing Costa Rica Institute of Technology Cartago Costa Rica School of Computer Engineering Costa Rica Institute of Technology Cartago Costa Rica School of Software Engineering & School of Computing Universidad CENFOTEC & Costa Rica Institute of Technology San José & Cartago Costa Rica School of Mathematics Costa Rica Institute of Technology Cartago Costa Rica

ISBN: (数字)9798331542788

ISBN: (纸本)9798331542795

Computational thinking is a high-level skill that involves both critical and creative thinking. This research proposes an advanced method for analyzing source code to assess computational thinking skills in elementary school students. We present a large-scale approach that examines source code from various programming exercises using a generalized abstract syntax tree, enabling language-independent analysis. Clustering techniques are applied to identify different levels of computational thinking development in areas such as parallelism, data representation, abstraction and decomposition, control flow, and programming structure. The results of the automated assessment provide insights into students' computational thinking abilities across diverse populations, highlighting opportunities for improving the educational framework.

关键词： Codes Source coding Data visualization Syntactics Parallel processing Pattern recognition software measurement engineering education Programming profession

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：