检索结果-内蒙古大学图书馆

Blockchain and signcryption enabled asynchronous federated learning framework in fog computing

Digital Communications and Networks 2025年第2期11卷 442-454页

作者： Zhou Zhou Youliang Tian Jinbo Xiong Changgen Peng Jing Li Nan Yang State Key Laboratory of Public Big Data College of Computer Science and TechnologyGuizhou UniversityGuiyang 550025China Fujian Provincial Key Laboratory of Network Security and Cryptology College of Computer and Cyber SecurityFujian Normal UniversityFuzhou 350117China Guangxi Key Laboratory of Cryptography and Information Security Guilin University of Electronic TechnologyGuilin 541004China

Federated learning combines with fog computing to transform data sharing into model sharing,which solves the issues of data isolation and privacy disclosure in fog ***,existing studies focus on centralized single-layer aggregation federated learning architecture,which lack the consideration of cross-domain and asynchronous robustness of federated learning,and rarely integrate verification mechanisms from the perspective of *** address the above challenges,we propose a Blockchain and Signcryption enabled Asynchronous Federated Learning(BSAFL)framework based on dual aggregation for cross-domain *** particular,we first design two types of signcryption schemes to secure the interaction and access control of collaborative learning between ***,we construct a differential privacy approach that adaptively adjusts privacy budgets to ensure data privacy and local models'availability of intra-domain ***,we propose an asynchronous aggregation solution that incorporates consensus verification and elastic participation using ***,security analysis demonstrates the security and privacy effectiveness of BSAFL,and the evaluation on real datasets further validates the high model accuracy and performance of BSAFL.

关键词： Blockchain Signcryption Federated learning Asynchronous Fog computing

来源：评论

学校读者我要写书评

暂无评论

Smart AI-Based Delivery Robot

Smart AI-Based Delivery Robot

引用

2025 IEEE International Conference on Computational, Communication and Information Technology, ICCCIT 2025

作者： Vemulapalli, Lavanya Charitha, Kolanti Harshitha, Kuralla Durga Lakshmi Sri, Vallepu Udaya Velagapudi Ramakrishna Siddhartha Engineering College Dept.of Computer Science and Engineering Vijayawada India Velagapudi Ramakrishna Siddhartha Engineering College Dept.of Artificial Intelligence and Data Science Vijayawada India

ISBN: (纸本)9798331512965

Humans are responsible for the majority of delivering the packets over the area of a network. When compared to classic delivery system the Autonomous Delivery Robot(ADR's) is meant to deliver the goods instead of a human involvement. The delivery robots are a contact-less and alternative to human interaction. At their core, we design a budget-friendly prototype for an Autonomous Delivery Robot (ADR) that can deliver packages to their destinations safely. This robot uses a GPS (like a car's navigation system) to find its way around. We use a special kind of computer brain called Artificial Intelligence to understand the best way to go. Once delivery robot know the right path, it follows until it reaches to the final stop. The robot keeps packets secure by putting them in a special container that needs a secret code to open. The secret code to open the container can be sent to the customer mobile when they order something. Once the robot delivers your package, it can find its own way back to where it started. © 2025 IEEE.

关键词： Budget control

来源：评论

学校读者我要写书评

暂无评论

FedFLD: Heterogeneous Federated Learning via Forget-Less Distillation

FedFLD: Heterogeneous Federated Learning via Forget-Less Dis...

引用

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

作者： Yi, Xiaoyang Zhang, Jian Chen, Jing Bao, Yuru Xing, Lingkai College of Cyber Science Nankai University China College of Computer Science Nankai University China Tianjin Key Laboratory of Network and Data Security Technology Tianjin China

ISBN: (纸本)9798350368741

Federated learning, as a distributed machine learning paradigm, enhances privacy protection but faces the challenge of heterogeneity. data-free knowledge distillation (DFKD) methods attempt to overcome this challenge by using a generator to synthesize samples for fine-tuning a global model. However, these methods often suffer from significant shifts in output distribution, leading to catastrophic forgetting. To tackle these issues, we propose FedFLD, a novel federated forget-less distillation framework that mitigates catastrophic forgetting in DFKD while addressing the heterogeneity challenge. Specifically, FedFLD guides the generator's training from three key aspects and constrains the output distribution with an elastic weight consolidation penalty term. By synthesizing diverse samples from different perspectives through additional generator updates, FedFLD facilitates effective knowledge distillation from local models to the global model. Additionally, the global model is further optimized via the heterogeneity fine-tuning process, mitigating the bias from heterogeneity and resulting in a more expressive and robust model. © 2025 IEEE.

关键词： Catastrophic Forgetting data-free Knowledge Distillation Federated learning Heterogeneity

来源：评论

学校读者我要写书评

暂无评论

Correspondence Learning via Group-wise Motion Enhancement 14

Correspondence Learning via Group-wise Motion Enhancement

引用

14th International Conference on Information Technology in Medicine and Education, ITME 2024

作者： Zhang, Yifan Kang, Zheng Lin, Xiaolang Lai, Taotao Lin, Lihui Li, Zuoyong Fuzhou University College of Computer and Data Science Fuzhou China Fujian Agriculture and Forestry University College of Computer and Information Sciences Fuzhou China Fujian University of Technology School of Transportation Fuzhou China Minjiang University School of Computer and Big Data Fuzhou China Wuyi University School of Mathematics and Computer Science Nanping China

ISBN: (纸本)9798350356670

The outlier removal methods are usually based on Multi-Layer Perceptron (MLP) for capturing context, which neglect the underlying motion information in images. Recently, CNN-based methods have attempted to address this issue by estimating motion fields and have made significant progress. However, CNNs tend to diminish the discontinuities of motion fields, making it difficult for the network to capture multiple motion patterns in complex scenes. Therefore, we propose a Group-wise Motion Enhancement (GME) block that groups motion fields along the channel dimension to learn multiple motion patterns. This enables the network to identify inliers under different motion patterns more accurately. Experiments show that compared with several existing methods, our method achieves advanced performance on the relative pose estimation task. © 2024 IEEE.

关键词： Motion estimation

来源：评论

学校读者我要写书评

暂无评论

A Smart Air Pollutants Monitoring System Using IOT Technologies

A Smart Air Pollutants Monitoring System Using IOT Technolog...

引用

2023 International Conference on Intelligent Systems for Communication, IoT and Security, ICISCoIS 2023

作者： Pyingkodi, M. Blessing, N R Wilfred Thenmozhi, K. Kanimozhi, K. Pavithira, L. Balakrishnan, P K Sandhya Kongu Engineering College Department of Computer Applications Erode India IT Department Oman Kristu Jayanti College Department of Computer Science Bengaluru India Sri Krishna Adithya College of Arts and Science Department of Information Technology Coimbatore India PSG College of Arts and Science College Department of Computer Science with Data Analytics Coimbatore India CMS College of Science and Commerce Department of Computer Science Coimbatore India

ISBN: (纸本)9798350335835

On international basis, one in all our century's essential problems is air pollutants. Every year, the range of individuals killed via way of means of pollutants climbs, with air pollutants being the main cause. The Internet of Things (IoT) is used worldwide community of 'inventive gadgets' which could sense, and join to, and have interaction with different humans and systems. The solution is furnished via way of means of a sensor-primarily based totally hardware module that can be deployed anywhere. These gadgets, which can be hooked up on mild poles, homes, and different structures, wirelessly speak air pleasant facts to a far flung server. The microprocessor collects facts from IoT sensing gadgets and sends it to a cloud server for processing. These facts can be used to expecting while the vicinity is maximum polluted. The machine additionally gives air pleasant statistics through a cell application, permitting clients to choose alternatives wherein the air pleasant is great. It is likewise a totally less expensive device. © 2023 IEEE.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

RLDR: Reinforcement Learning-Based Fast data Recovery in Cloud-of-Clouds Storage Systems

引用

IEEE Transactions on Cloud Computing 2025年第2期13卷 526-543页

作者： Shen, Jiajie Wu, Bochun Wang, Maoyi Zou, Sai Cui, Laizhong Ni, Wei Fudan University Shanghai200433 China Guizhou University College of Big Data and Information Engineering Guiyang550025 China Shenzhen University College of Computer Science and Software Engieering Shenzhen518060 China Commonwealth Scientific and Industrial Research Organization Data61 EastwoodNSW2122 Australia

Cloud-of-clouds storage systems are widely used in online applications, where user data are encrypted, encoded, and stored in multiple clouds. When some cloud nodes fail, the storage systems can reconstruct the lost data and store it in the substitute nodes. It is a challenge to reduce the latency of data recovery to ensure data reliability. In this paper, we adopt a Reinforcement Learning-based data Recovery (RLDR) approach to reduce the regeneration time. By employing the Monte-Carlo method, our approach can construct the tree-topology-based regeneration process, a.k.a. regeneration tree, to effectively reduce the regeneration time. Through rigorous analysis, we apply the information flow graph to optimize the inter-cloud traffic for a given regeneration tree. To verify the merit of RLDR, We conduct extensive experiments on real-world traces. Experiments demonstrate that RLDR can significantly accelerate the regeneration process. Specifically, RLDR can reduce the regeneration time by up to 92% and increase the throughput by up to twelve-fold, compared to the prior art. © 2013 IEEE.

关键词： Cloud storage

来源：评论

学校读者我要写书评

暂无评论

PRESS: Defending Privacy in Retrieval-Augmented Generation via Embedding Space Shifting

PRESS: Defending Privacy in Retrieval-Augmented Generation v...

引用

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

作者： He, Jiaming Liu, Cheng Hou, Guanyu Jiang, Wenbo Li, Jiachen School of Computer Science and Engineering University of Electronic Science and Technology of China China College of Computer Science and Cyber Security Oxford Brookes College Chengdu University of Technology China School of Data Science The Chinese University of Hong Kong Shenzhen China School of Computer Science and Artificial Intelligence Wuhan University of Technology China

ISBN: (纸本)9798350368741

Retrieval-augmented generation (RAG) expands the capabilities of large language models (LLMs) in various applications by integrating relevant information retrieved from external data sources. However, the RAG systems are exposed to substantial privacy risks during the information retrieval process, leading to potential data leakage of private information. In this work, we present a Privacy-preserving Retrieval-augmented generation via Embedding Space Shifting (PRESS), systematically exploring how to protect privacy in RAG systems. Specifically, we first conduct proximal policy optimization (PPO) based training on pre-trained language models to generate target training samples. Then we employ a purposive shift fine-tuning on the text embedding model with the generated samples for guiding the RAG system to map potential privacy leaking queries to safe target in embedding space. Extensive experimental results on representative models and datasets demonstrate that our protection method achieves high defense performance with high efficiency while keeping the normal functionality of the RAG system. © 2025 IEEE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Coins Game - A Novel Decision Problem and Its Solving Using Reinforcement Learning 4

Coins Game - A Novel Decision Problem and Its Solving Using ...

引用

4th Asia Conference on Information Engineering, ACIE 2024

作者： Chen, Bo Lin, Ziqing Chen, Guohong Fuzhou University College of Computer and Data Science Fuzhou China Fuzhou University School of Economics and Management Fuzhou China

ISBN: (纸本)9798350371925

Recent years have witnessed continuous optimization and innovation of reinforcement learning algorithms. Games, as a key application paradigm, have been widely employed to develop superior reinforcement learning models. Different game environments present diverse challenges to reinforcement learning agents;however, mainstream gaming paradigms have not yet specifically addressed issues such as variable action spaces and varying tasks. In this regard, this paper introduces a two-player adversarial game with a configurable player action space. The game allows for the diversification of task challenges by configuring opponent strategies. Additionally, we propose a reinforcement learning method to facilitate the decision-making of the AI player (the agent) in the game. The inclusion of an action masking algorithm enables effective handling of variable action space issues. Experimental results indicate that the decision-making behavior of the agent adjusts with changes in opponent behavior and continuously improves with policy updates. The trained agent exhibits impressive performance in this game, it shows that the proposed method could serve as a baseline for the decision-making in the novel game, and a robust foundation for further research and applications is provided. © 2024 IEEE.

关键词： Decision making

来源：评论

学校读者我要写书评

暂无评论

Secure and Low-Complexity Medical Image Exchange Based on Compressive Sensing and LSB Audio Steganography

引用

Informatica (Slovenia) 2023年第6期47卷 65-74页

作者： Al-Chaab, Wafaa Abduljabbar, Zaid Ameen Abood, Enas Wahab Nyangaresi, Vincent Omollo Mohammed, Hussein M. Ma, Junchao Department of Mathematics College of Science University of Basrah Basrah61004 Iraq Department of Computer Science College of Education for Pure Sciences University of Basrah Basrah61004 Iraq Technical Computer Engineering Department Al-Kunooze University College Basrah61001 Iraq Huazhong University of Science and Technology Shenzhen Institute Shenzhen430074 China Jaramogi Oginga Odinga University of Science & Technology Bondo40601 Kenya Department of Computer Science Shatt Al-Arab University College Basra61001 Iraq College of Big Data and Internet Shenzhen Technology University Shenzhen518118 China

Patients’ data constitutes the vast majority of information exchanged over the Internet. The magnitude of maintaining its security and privacy directly correlates to the importance and privacy of the patients themselves. The manipulation of this data negatively impacts the patient's life and treatment, thus the researchers were inclined to find solutions for this confidentiality challenge by concealing the data to prevent unauthorized access. The process of data concealment is more complex when dealing with large amounts of data. Often, the proportions of the medical image are sizable, which creates challenges when compressing data while maintaining high accuracy. In this paper, a system for securing and compressing medical images based on the Compressive Sensing (CS) principle is presented. The medical image is divided into 8×4 sub-matrices that are multiplied by a 3×8 sensor matrix consisting of Gaussian random numbers. The proposed solution reduces the image's original size by about 30% and conceals it as a random distribution inside an audio (wav) file for more security, using the LSB technique for low complexity. For reconstruction, the sensed image is multiplied by the pseudo-inverse of Moore-Penrose. The statistical analysis proved the efficacy of the system in compression and recovery with reduced cost and time consumption, combined with reduced distortion of the cover file;it was also judged to be increasingly efficient compared to previous research. © 2023 Slovene Society Informatika. All rights reserved.

关键词： Compressed sensing

来源：评论

学校读者我要写书评

暂无评论

Enhance Efficient of Support Vector Machine with CNN and PSO for solving Handwritten Digits Recognition 28

Enhance Efficient of Support Vector Machine with CNN and PSO...

引用

28th International computer science and Engineering Conference, ICSEC 2024

作者： Janprasit, Siwach Punkong, Narong Ratanavilisagul, Chiabwoot Kosolsombat, Somkiat Faculty of Applied Science Department of Computer and Information Science Bangkok Thailand Digital Technology for Business Faculty of Management Science Kanchanaburi Thailand Faculty of Applied Science Department of Computer and Information Sciences Bangkok Thailand Thammasat University Data Science and Innovation College of Interdisciplinary Studies Thailand

ISBN: (纸本)9798350366860

handwritten digit recognition is a crucial task in various fields such as postal mail sorting, bank check processing, and digitizing handwritten documents. This research aims to compare the effectiveness of using Convolutional Neural Network (CNN) for feature extraction combined with Support Vector Machine (SVM) for classification, against the traditional method of using Discrete Wavelet Transform (DWT) and Discrete Cosine Transform (DCT) for feature extraction. Additionally, Particle Swarm Optimization (PSO) is applied to optimize SVM parameters to further enhance model performance. The experiments are conducted on two datasets: The MNIST dataset and an individual handwriting dataset from 50 participants. The results indicate that the proposed method of using CNN for feature extraction and PSO-optimized SVM achieves superior accuracy compared to traditional DWT and DCT methods. Specifically, the proposed method achieved an accuracy of 99.05% on the MNIST dataset and 67.79% on the individual handwriting dataset, demonstrating its robustness and effectiveness in handwritten digit recognition. © 2024 IEEE.

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：