检索结果-内蒙古大学图书馆

Optimizing Wireless Resource Management and Synchronization in Digital Twin Networks

IEEE Internet of Things Journal 2025年

作者： Yu, Hanzhi Liu, Yuchen Yang, Zhaohui Sun, Haijian Chen, Mingzhe University of Miami Department of Electrical and Computer Engineering Frost Institute for Data Science and Computing Coral GablesFL33146 United States North Carolina State University Department of Computer Science RaleighNC27695 United States Zhejiang University College of Information Science and Electronic Engineering Hangzhou310027 China University of Georgia School of Electrical and Computer Engineering AthensGA30602 United States

In this paper, we investigate an accurate synchronization between a physical network and its digital network twin (DNT), which serves as a virtual representation of the physical network. The considered network includes a set of base stations (BSs) that must allocate its limited spectrum resources to serve a set of users while also transmitting its partially observed physical network information to a cloud server to generate the DNT. Since the DNT can predict the physical network status based on its historical status, the BSs may not need to send their physical network information at each time slot, allowing them to conserve spectrum resources to serve the users. However, if the DNT does not receive the physical network information of the BSs over a large time period, the DNT's accuracy in representing the physical network may degrade. To this end, each BS must decide when to send the physical network information to the cloud server to update the DNT, while also determining the spectrum resource allocation policy for both DNT synchronization and serving the users. We formulate this resource allocation task as an optimization problem, aiming to maximize the total data rate of all users while minimizing the asynchronization between the physical network and the DNT. The formulated problem is challenging to solve by traditional optimization methods, as each BS can only observe a partial physical network, making it difficult to find an optimal spectrum allocation strategy for the entire network. To address this problem, we propose a method based on the gated recurrent units (GRUs) and the value decomposition network (VDN). The GRU component allows the DNT to predict future status using the historical data, effectively updating itself when the BSs do not transmit the physical network information. The VDN algorithm enables each BS to learn the relationship between its local observation and the team reward of all BSs, allowing it to collaborate with others in determining whe

关键词： Resource allocation

来源：评论

学校读者我要写书评

暂无评论

Enhancing Jailbreak Attacks via Compliance-Refusal-Based Initialization

arXiv

引用

arXiv 2025年

作者： Levi, Amit Himelstein, Rom Nemcovsky, Yaniv Mendelson, Avi Baskin, Chaim Department of Computer Science Technion - Israel Institute of Technology Department of Data and Decision Science Technion - Israel Institute of Technology School of Electrical and Computer Engineering Ben-Gurion University of The Negev

Jailbreak attacks aim to exploit large language models (LLMs) and pose a significant threat to their proper conduct;they seek to bypass models’ safeguards and often provoke transgressive behaviors. However, existing automatic jailbreak attacks require extensive computational resources and are prone to converge on suboptimal solutions. In this work, we propose Compliance Refusal Initialization (CRI), a novel, attack-agnostic framework that efficiently initializes the optimization in the proximity of the compliance subspace of harmful prompts. By narrowing the initial gap to the adversarial objective, CRI substantially improves adversarial success rates (ASR) and drastically reduces computational overhead—often requiring just a single optimization step. We evaluate CRI on the widely-used AdvBench dataset over the standard jailbreak attacks of GCG and AutoDAN. Results show that CRI boosts ASR and decreases the median steps to success by up to ×60. The project page, along with the reference implementation, is publicly available at . © 2025, CC BY-NC-SA.

关键词： Crime

来源：评论

学校读者我要写书评

暂无评论

Avoiding exp(Rmax) scaling in RLHF through Preference-based Exploration

arXiv

引用

arXiv 2025年

作者： Chen, Mingyu Chen, Yiding Sun, Wen Zhang, Xuezhou Department of Electrical & Computer Engineering Boston University United States Department of Computer Science Cornell University United States Faculty of Computing & Data Sciences Boston University United States

Reinforcement Learning from Human Feedback (RLHF) has emerged as a pivotal technique for large language model (LLM) alignment. This paper studies the setting of online RLHF and focus on improving sample efficiency. All existing algorithms in online RLHF, whether doing passive exploration or active exploration, suffer from a sample complexity that scales exponentially with the scale of the reward function. This fundamental limitation hinders their effectiveness in scenarios with heavily skewed preferences, e.g. questions with a unique correct solution. To address this, we introduce Self-Exploring Preference-Incentive Online Preference Optimization (SE-POPO), an online RLHF algorithm that for the first time achieves a sample complexity that scales polynomially with the reward scale, answering an open problem raised by Xie et al. (2024). Theoretically, we demonstrate that the sample complexity of SE-POPO dominates that of existing exploration algorithms. Empirically, our systematic evaluation confirms that SE-POPO is more sample-efficient than both exploratory and non-exploratory baselines, in two primary application scenarios of RLHF as well as on public benchmarks, marking a significant step forward in RLHF algorithm design. The code is available at https://***/MYC000801/SE-POPO. © 2025, CC BY.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

A Hybrid GAN-CNN Framework for Low-Light Image Enhancement Combining Structural Noise Reduction and Perceptual Quality

A Hybrid GAN-CNN Framework for Low-Light Image Enhancement C...

引用

2025 International Conference on Multi-Agent Systems for Collaborative Intelligence, ICMSCI 2025

作者： Gobinath, R. Boopathi, R. Vijayalakshmi, S. Disney, D. Anu Priyadharshini, P. Nirmala Maheshwari, K. E.G.S. Pillay Engineering College Nagapattinam611002 India Karpaga Vinayaga College of Engineering and Technology Department of Electrical and Electronic Engineering Tamilnadu 603308 India Sona College of Technology Department of Electronics and communication Engineering Tamilnadu Salem636005 India Sathyabama Institute of Science and Technology Department of Computer Science and Engineering Chennai600119 India Adithya Institute of Technology Department of Information Technology Coimbatore641107 India Sir Issac Newton College of Engineering and Technology Department of Artificial Intelligence and Data Science Nagapattinam611 102 India

ISBN: (纸本)9798331509828

Low light region is one of the difficult scenarios for surveillance system, especially noisy, low-quality images may lead to poor object detection and scene analysis. To tackle this problem, we introduce a hybrid GAN - CNN framework for low light image denoising and quality enhancement. We develop a framework that combines the strengths of Generative Adversarial Networks (GANs) for photorealistic image generation and Convolutional Neural Networks (CNNs) for robust noise reduction. First denoising and feature extraction is done by the CNN module, and the GAN part further refines the denoised output to improve the perceptual quality of the generated images. We train and validate the proposed model in standard low light image dataset and show superior PSNR, SSIM, and visual clarity compared to state of art methods. We show that the hybrid GAN-CNN model successfully trades between image denoising accuracy and visual realism, thereby providing a viable alternative for enhancing surveillance performance in lowlight conditions. Keywords: Image denoising, GAN-CNN hybrid, low-light enhancement, PSNR, SSIM, surveillance systems. © 2025 IEEE.

关键词： Gallium nitride

来源：评论

学校读者我要写书评

暂无评论

Advances and Open Challenges in Federated Foundation Models

引用

IEEE Communications Surveys and Tutorials 2025年

作者： Ren, Chao Yu, Han Peng, Hongyi Tang, Xiaoli Zhao, Bo Yi, Liping Tan, Alysa Ziying Gao, Yulan Li, Anran Li, Xiaoxiao Li, Zengxiang Yang, Qiang School of Electrical Engineering and Computer Science Kth Royal Institute of Technology Sweden College of Computing and Data Science Nanyang Technological University Singapore College of Computer Nankai University China School of Medicine Yale University United States Digital Research Institute of Enn Group Langfang China Department of Electrical and Computer Engineering The University of British Columbia VancouverBC Canada Department of Computer Science and Engineering Hong Kong University of Science and Technology Hong Kong WeBank Shenzhen China

The integration of Foundation Models (FMs) with Federated Learning (FL) presents a transformative paradigm in Artificial Intelligence (AI). This integration offers enhanced capabilities, while addressing concerns of privacy, data decentralization and computational efficiency. This paper provides a comprehensive survey of the emerging field of Federated Foundation Models (FedFM), elucidating their synergistic relationship and exploring novel methodologies, challenges, and future directions that the FL research field needs to focus on in order to thrive in the age of FMs. A systematic multi-tiered taxonomy is proposed, categorizing existing FedFM approaches for model training, aggregation, trustworthiness, and incentivization. Key challenges, including how to enable FL to deal with high complexity of computational demands, privacy considerations, contribution evaluation, and communication efficiency, are thoroughly discussed. Moreover, this paper explores the intricate challenges of communication, scalability and security inherent in training/fine-tuning FMs via FL. It highlights the potential of quantum computing to revolutionize the processes of training, inference, optimization and security. This survey also introduces the implementation requirement of FedFM and some practical FedFM applications. It highlights lessons learned with a clear understanding of our findings for FedFM. Finally, this survey not only provides insights into the current state and challenges of FedFM, but also offers a blueprint for future research directions, emphasizing the need for developing trustworthy solutions. It serves as a foundational guide for researchers and practitioners interested in contributing to this interdisciplinary and rapidly advancing field. © 2025 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Machine Learning for Predicting Energy Consumption in Smart Buildings Using Artificial Adaptive Systems

Machine Learning for Predicting Energy Consumption in Smart ...

引用

2025 International Conference on Multi-Agent Systems for Collaborative Intelligence, ICMSCI 2025

作者： Kocherla, Raviteja Jency, C. Mohan, T. Mani Mallala, Balasubbareddy Manikandaprabu, P. Kanth, Tatiraju V. Rajani Mallareddy University Department of Computer Science & Engineering Telangana Hyderabad500043 India Dr. N.G.P. Institute of Technology Department of Artificial Intelligence and Data Science Coimbatore641048 India Annamacharya University Department of Mechanical Engineering Rajampet516126 India ChaitanyaBharathi Institute of Technology Department of Electrical and Electronics Engineering Telangana Hyderabad500075 India Sona College of Technology Department of Mechanical Engineering Salem636005 India Tvr Consulting Services Private Limited Telangana Hyderabad500040 India

ISBN: (纸本)9798331509828

This study investigates the use of Artificial Adaptive (AA) systems for predicting energy consumption in Smart Buildings, focusing on optimizing energy usage with sensors, actuators, and intelligent systems. Using the CRISP-DM methodology, the research incorporated statistical techniques like PCA and ARIMA models to extract relevant features and identify temporal relationships. Long Short-Term Memory (LSTM) networks were employed to refine prediction models, balancing neurons and training epochs to prevent overfitting or underfitting. The best model, utilizing the first Principal Component, achieved 74% precision. The findings highlight AA systems' potential in energy management, emphasizing the challenges of model precision and resource availability, while offering a foundation for future improvements in predictive modeling. © 2025 IEEE.

关键词： Smart homes

来源：评论

学校读者我要写书评

暂无评论

Hybrid Beamforming Optimization in Millimeter Wave MIMO Using Deep Deterministic Policy Gradient Based Deep Reinforcement Learning 3

Hybrid Beamforming Optimization in Millimeter Wave MIMO Usin...

引用

3rd International Conference on Integrated Circuits and Communication Systems, ICICACS 2025

作者： Riadhusin, Rame Sudhakar, K. Prakash, D. Bhanu Kumar, S. Senthil Rajasubha, J. College of Technical Engineering The Islamic University Department of Computers Techniques Engineering Najaf Iraq Nitte Meenakshi Institute of Technology Department of Artificial Intelligence and Data Science Bengaluru India Cvr College of Engineering Department of Electronics and Communication Engineering Hyderabad India New Prince Shri Bhavani College of Engineering and Technology Department of Electrical and Electronics Engineering Chennai India Dhanalakshmi Srinivasan College of Engineering Technology Department of Computer Science and Engineering Mamallapuram India

ISBN: (纸本)9798331508456

In recent years, the rapid growth of Hybrid Beamforming (HB) technology has developed rapidly to improve high spectrum efficiency and data transmission in wireless communication systems. Massive Multiple Input Multiple Output (MIMO) systems are integrated with HB to encourage wireless network performance by improving signal quality. However, combination of digital and analog beamforming increases hardware complexity and quantization errors in multiple streams. Hence, this research proposes Deep Reinforcement Learning (DRL)-based hybrid beamforming optimization framework to learn relevant Channel State Information (CSI) by learning complex non-linear relationships and beam forming weights dynamically. Initially, Deep Multiple Input Multiple Output (Deep MIMO) dataset which consists of channel measurements and capacity to receive and transmit signal processing is used. After that, preprocessing is performed using Amplitude Scaling (AS) for signal compression and Phase Normalization (PN) to reduce phase noise in multiple frequencies. Finally, Deep Deterministic Policy Gradient (DDPG)-based DRL model is used to perform HB optimization through adaptive learning policy by improving sum rate and spectral efficiency while reducing quantization errors. Proposed method results with the Sum-rate (5.35 bps/HZ), Accuracy (99.26%) and Inference time (0.03 sec) when compared with existing CNN-Bi-directional Long Short-Term Memory (CNNBiLSTM). © 2025 IEEE.

关键词： Beamforming

来源：评论

学校读者我要写书评

暂无评论

IoT-Enabled Smart Irrigation System with ABiLSTM for Optimized Water Management and Crop Yields 3

IoT-Enabled Smart Irrigation System with ABiLSTM for Optimiz...

引用

3rd International Conference on Integrated Circuits and Communication Systems, ICICACS 2025

作者： Chaturvedi, Abhay Rao, Praveena Venkata Maruti Prasad, S. Malini, T. Prabagar, S. Bharathi, K. Gla University Department of Electronics and Communication Engineering Mathura India Institute of Aeronautical Engineering Department of Civil Engineering Hyderabad India Aditya Institute of Technology & Management Department of Bs & H Tekkali India Sri Krishna College of Engineering and Technology Department of Electrical and Electronics Engineering Coimbatore India School of Advanced Computing Alliance University Department of Computer Science and Engineering Bangalore India Erode Sengunthar Engineering College Department of Agricultural Engineering Perundurai India

ISBN: (纸本)9798331508456

An efficient management of water resources in agriculture is now crucial in light of the growing global population and increased demand for water. Innovative solutions are needed to improve water efficiency since conventional irrigation methods are not sustainable anymore. Through the use of precision agriculture and the Internet of Things (IoT), smart irrigation has revolutionised water management, leading to increased crop yields. In order to improve the efficiency of decision-making, this research presents an Internet of Things (IoT) smart irrigation system that uses SA-BiLSTM (Self-Attention Bidirectional Long Short-Term Memory). Using Min-Max and Z-score normalisation approaches for feature scaling, the methodology includes data preparation, feature selection, and model training. Methods based on ranking, such as the Chi-Square Test, Pearson Correlation Coefficient (PCC), and Mutual Information (MI), were used to prioritise feature selection. Classification and prediction testing showed that the proposed SA-BiLSTM model outperformed state-of-the-art models like HGAT and Linear SVM, with an astounding 95.63% accuracy. Improved resource allocation and irrigation effectiveness, leading to increased agricultural output, are highlighted by the results, which highlight the effectiveness of advanced machine learning approaches in precision agriculture. © 2025 IEEE.

关键词： Resource allocation

来源：评论

学校读者我要写书评

暂无评论

Joint Task Partitioning and Resource Allocation in UAV-Enabled Vehicular Edge computing Based on Deep Reinforcement Learning

引用

IEEE Internet of Things Journal 2025年第11期12卷 15453-15466页

作者： Liang, Hongbin Zhang, Han Ale, Laha Hong, Xintao Wang, Lei Jia, Qiong Zhao, Dongmei Southwest Jiaotong University School of Transportation and Logistics National United Engineering Laboratory of Integrated and Intelligent Transportation National Engineering Laboratory of Integrated Transportation Big Data Application Technology Chengdu611756 China Southwest Jiaotong University School of Computing and Artificial Intelligence Chengdu611756 China Southwest Jiaotong University School of Information Science and Technology Chengdu611756 China Chinese Academy of Sciences Laboratory of Noise and Vibration Research Institute of Acoustics Beijing100190 China McMaster University Department of Electrical and Computer Engineering HamiltonONL8S 4K1 Canada

Vehicle Edge computing (VEC) leverages compact cloud computing at the mobile network edge to meet the processing and latency needs of vehicles. By bringing computation closer to the vehicles, VEC reduces data transmission, minimizes latency, and boosts performance for compute-intensive applications. However, during peak hours of urban road traffic, the scarce computational resources available at edge servers could pose challenges in fulfilling the processing needs of vehicles. Introducing Unmanned Aerial Vehicles (UAVs) as supplementary edge computing nodes could significantly mitigate the aforementioned issue. In this paper, we propose a flexible edge computing framework in which a fleet of UAVs function as mobile computational service providers, offering computation offloading services to multiple vehicles. We design and optimize a computation offloading model for the UAV-enabled vehicle edge computing environment. The proposed model tackles the task offloading challenge, aiming to optimize UAV revenue and task processing efficiency while considering the constraints of UAVs' restricted computational power and energy resources. Towards this end, our model jointly considers two key factors: task partitioning and computational resource allocation. To tackle the challenges posed by the aforementioned non-convex optimization problem, we construct a Markov Decision Process (MDP) model for the multi-UAV-enabled mobile edge computing system and introduce an innovative Multi-Agent Deep Reinforcement Learning (MADRL) framework addressing the decision-making challenge represented by MDP model. Comprehensive simulation outcomes illustrate that our devised task offloading technique outperforms other optimization methods. © 2014 IEEE.

关键词： Convex optimization

来源：评论

学校读者我要写书评

暂无评论

Securing Consumer IoT Swarms Using Graph Transformers and SRAM for Firmware Attestation

Securing Consumer IoT Swarms Using Graph Transformers and SR...

引用

Consumer Communications and Networking Conference, CCNC IEEE

作者： Varun Kohli Bhavya Kohli Muhammad Naveed Aman Biplab Sikdar Department of Electrical and Computer Engineering National University of Singapore Singapore Center for Machine Intelligence and Data Science Indian Institute of Technology Bombay India School of Computing University of Nebraska-Lincoln USA

ISBN: (数字)9798331508050

ISBN: (纸本)9798331508067

Consumer Internet of Things (IoT) networks have gained widespread popularity due to their convenience, automation, and security provisions in personal and home environments. Ubiquitous resource-constrained devices, however, are plagued with security issues that often arise from firmware-related issues and their propagated effects. While various studies on firmware attestation are available, they require firmware copies, specific hardware, and complex computation on the IoT device. This paper presents a study on the application of Graph Transformer Networks (GTN) in verifying the firmware integrity of consumer IoT swarms using SRAM as an attestation feature. The proposed method achieves an overall 0.99 accuracy on authentic samples from development and physical twin networks, 0.99 on malware, and 0.97 on propagated misbehavior at a $\sim 10^{-4}$ second inference latency on a laptop CPU.

关键词： Portable computers Accuracy Random access memory Transformers Malware Graph neural networks Hardware Internet of Things Security Microprogramming

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：