检索结果-内蒙古大学图书馆

12th IEEE International Conference on Identification, Information and Knowledge in the Internet of Things, IIKI 2024

作者： Alsharif, Maram Rawat, Danda B. Department of Electrical Engineering & Computer Science WashingtonDC20059 United States

ISBN: (纸本)9798331510633

Security solutions based on machine learning enabled Intrusion detection systems (ML-IDS) have been popular for IoT edge device security. The cloud support has been considered to leverage heavy workloads off IoT edge devices such as learning, model fitting and updating. In this paper, we apply ML model optimization techniques as 3 distinct key roles of the cloud layer in Machine Learning as a Service (MaaS) architecture. In order to influence the performance of IoT edge devices, the optimization would require: a) device benchmarking that quantifies the available computing resources and network traffic for the edge device;b) feature weighing that determines the contribution of each extracted feature to the overall ML-IDS operation;and c) dimensionality reduction of important features in the training dataset. We examine feature importance and dimensionality reduction techniques on 3 public IDS datasets and present how they affect ML model performance. © 2024 IEEE.

关键词： Cloud security

来源：评论

学校读者我要写书评

暂无评论

Beyond Human Vision: The Role of Large Vision Language Models in Microscope Image Analysis

Beyond Human Vision: The Role of Large Vision Language Model...

引用

2024 IEEE International Conference on Big Data, BigData 2024

作者： Verma, Prateek Van, Minh-Hao Wu, Xintao University of Arkansas Department of Electrical Engineering and Computer Science Fayetteville United States

ISBN: (纸本)9798350362480

Vision language models (VLMs) such as LLaVA, ChatGPT-4, and Gemini have recently emerged and gained the spotlight for their ability to comprehend the dual modality of image and textual data showing impressive performance on tasks such as natural image captioning, visual question answering, and spatial reasoning. Additionally, a universal segmentation model by Meta AI, Segment Anything Model (SAM) shows unprecedented performance at isolating objects from unforeseen images. Because medical experts, biologists, and materials scientists routinely examine microscopy or medical images in conjunction with textual information in the form of captions, literature, or reports, and draw conclusions of great importance and merit, it is essential to evaluate their performance on these images. In this study, we charge ChatGPT, LLaVA, Gemini, and SAM quantitatively with classification, segmentation and counting tasks. We observed that ChatGPT and Gemini were impressively able to comprehend the visual features in microscopy images, while SAM was quite capable at isolating artifacts in a general sense. However, the performance was not close to that of a domain expert - the models were readily encumbered by the introduction of impurities, defects, object overlaps and diversity present in the images. © 2024 IEEE.

关键词： biology electron microscopy images materials science vision language models zero-shot evaluation

来源：评论

学校读者我要写书评

暂无评论

A Novel Blockchain-based System for Service Quality Improvement in Multi-Tenant O-RANs

A Novel Blockchain-based System for Service Quality Improvem...

引用

2024 IEEE Global Communications Conference, GLOBECOM 2024

作者： Zhang, Jiarui Shang, Xiaojun Zeng, Yiming Yang, Yuanyuan Stony Brook University Department of Electrical and Computer Engineering United States University of Texas Department of Computer Science and Engineering Arlington United States

ISBN: (纸本)9798350351255

Open Radio Access Networks (O-RANs) are transforming the landscape of telecommunications to better performance and higher cost-efficiency by enabling network operators to integrate diverse vendor components. Nevertheless, the involvement of multiple Network Service Providers (NSPs) and Mobile Network Operators (MNOs) also brings new challenges in the management of computation and network resources. To resolve this challenge, we propose a blockchain-based framework to guarantee secure, transparent, and decentralized resource allocation in O-RAN systems. Our resource allocation mainly considers the tradeoff of cost and service quality. Our design facilitates real-time adjustments to resource distribution based on dynamic network conditions and incorporates user feedback to optimize service quality continuously. By integrating a Proof-of-Reputation (PoR) consensus mechanism, the framework enhances the reliability and integrity of transactions among competing vendors without central oversight. We evaluate the performance of our design through extensive simulations, which demonstrate significant improvements over the baselines in resource utilization and service delivery across various network scenarios. © 2024 IEEE.

关键词： Radio access networks

来源：评论

学校读者我要写书评

暂无评论

HETEROGENEOUS PERSONALIZED FEDERATED LEARNING BY LOCAL-GLOBAL UPDATES MIXING VIA CONVERGENCE RATE 12

HETEROGENEOUS PERSONALIZED FEDERATED LEARNING BY LOCAL-GLOBA...

引用

12th International Conference on Learning Representations, ICLR 2024

作者： Jiang, Meirui Le, Anjie Li, Xiaoxiao Dou, Qi Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong Department of Electrical and Computer Engineering The University of British Columbia Canada

Personalized federated learning (PFL) has emerged as a promising technique for addressing the challenge of data heterogeneity. While recent studies have made notable progress in mitigating heterogeneity associated with label distributions, the issue of effectively handling feature heterogeneity remains an open question. In this paper, we propose a personalization approach by Local-Global updates Mixing (LG-Mix) via Neural Tangent Kernel (NTK)-based convergence. The core idea is to leverage the convergence rate induced by NTK to quantify the importance of local and global updates, and subsequently mix these updates based on their importance. Specifically, we find the trace of the NTK matrix can manifest the convergence rate, and propose an efficient and effective approximation to calculate the trace of a feature matrix instead of the NTK matrix. Such approximation significantly reduces the cost of computing NTK, and the feature matrix explicitly considers the heterogeneous features among samples. We have theoretically analyzed the convergence of our method in the over-parameterize regime, and experimentally evaluated our method on five datasets. These datasets present heterogeneous data features in natural and medical images. With comprehensive comparison to existing state-of-the-art approaches, our LG-Mix has consistently outperformed them across all datasets (largest accuracy improvement of 5.01%), demonstrating the outstanding efficacy of our method for model personalization. Code is available at https://***/med-air/HeteroPFL. © 2024 12th International Conference on Learning Representations, ICLR 2024. All rights reserved.

关键词： Mixing

来源：评论

学校读者我要写书评

暂无评论

ACHIEVING SAMPLE AND COMPUTATIONAL EFFICIENT REINFORCEMENT LEARNING BY ACTION SPACE REDUCTION VIA GROUPING 12

ACHIEVING SAMPLE AND COMPUTATIONAL EFFICIENT REINFORCEMENT L...

引用

12th International Conference on Learning Representations, ICLR 2024

作者： Li, Yining Ju, Peizhong Shroff, Ness Department of Electrical and Computer Engineering United States Department of Computer Science and Engineering The Ohio State University ColumbusOH43210 United States

Reinforcement learning often needs to deal with the exponential growth of states and actions when exploring optimal control in high-dimensional spaces (often known as the curse of dimensionality). In this work, we address this issue by learning the inherent structure of action-wise similar MDP to appropriately balance the performance degradation versus sample/computational complexity. In particular, we partition the action spaces into multiple groups based on the similarity in transition distribution and reward function, and build a linear decomposition model to capture the difference between the intra-group transition kernel and the intra-group rewards. Both our theoretical analysis and experiments reveal a surprising and counter-intuitive result: while a more refined grouping strategy can reduce the approximation error caused by treating actions in the same group as identical, it also leads to increased estimation error when the size of samples or the computation resources is limited. This finding highlights the grouping strategy as a new degree of freedom that can be optimized to minimize the overall performance loss. To address this issue, we formulate a general optimization problem for determining the optimal grouping strategy, which strikes a balance between performance loss and sample/computational complexity. We further propose a computationally efficient method for selecting a nearly-optimal grouping strategy, which maintains its computational complexity independent of the size of the action space. © 2024 12th International Conference on Learning Representations, ICLR 2024. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Binary Logistic Regression based Intrusion Detection System for CAN bus Security 12

Binary Logistic Regression based Intrusion Detection System ...

引用

12th IEEE International Conference on Identification, Information and Knowledge in the Internet of Things, IIKI 2024

作者： Alfardus, Asma Rawat, Danda B. Department of Electrical Engineering and Computer Science WashingtonDC20059 United States

ISBN: (纸本)9798331510633

The complex distributed systems installed in vehicles represent the cutting edge of the automotive industry. Electronic control units communicate with each other by sending and receiving messages over a well-known protocol called the Controller Area Network (CAN) bus system. Car Broadcast is a new, but insecure, way to communicate between external electronic devices. However, today's vehicles are on the brink of security because the CAN network lacks a secure authentication and authorization mechanism. The rise in cyber attacks such as spoofing, spoofing and most commonly denial of service attacks is the result of uncertain measures in the CAN bus network. Although many intrusion detection systems have been developed to provide more secure communication in the vehicle, CAN is still far from being the most secure communication protocol. Since cyber attacks can come from a little-known or completely unknown source, it is essential to take a probabilistic approach based on previous observations from previous attacks. Therefore, we propose a new intrusion detection system that uses binary logistic regression (BLR) to detect and mitigate attacks on a CAN bus network. Binary logistic regression is a very popular predictive model that is widely used in various fields. In binary logistic regression, data is first analyzed, then the probability of individual events is estimated by observing previous data, and then a binary classification model is created. An evaluation of the well-known Nsl-kdd and Kdd-99 datasets shows that our proposed method has a dominant overall performance. The final detection rate is 099.031% using Nsl-kdd with a positive rate as low as 00.073% and the rate of detection is 099.043% using kdd-99 with a positive false rate as low as 00.046%˙ Specifically, to detect denial of service (DoS) attacks, the proposed system achieved a detection rate of 099.061% and 099.098% in Nsl-kdd and kdd-99 dataset respectively. Comparative evaluation confirmed that BLR i

关键词： Automotive industry

来源：评论

学校读者我要写书评

暂无评论

An Optimal DPM Based Energy-Aware Task Scheduling for Performance Enhancement in Embedded MPSoC

引用

computers, Materials & Continua 2023年第1期74卷 2097-2113页

作者： Hamayun Khan Irfan Ud Din Arshad Ali Mohammad Husain Department of Electrical Engineering Faculty of Engineering&Technology Superior University Lahore54000Pakistan Department of Computer Science Faculty of Computer Science&IT Superior University Lahore54000Pakistan Faculty of Computer and Information Systems Islamic University of MadinahAl Madinah Al Munawarah42351Saudi Arabia

Minimizing the energy consumption to increase the life span and performance of multiprocessor system on chip(MPSoC)has become an integral chip design issue for multiprocessor *** performance measurement of computational systems is changing with the advancement in *** to shrinking and smaller chip size power densities onchip are increasing rapidly that increasing chip temperature in multi-core embedded *** operating speed of the device decreases when power consumption reaches a threshold that causes a delay in complementary metal oxide semiconductor(CMOS)circuits because high on-chip temperature adversely affects the life span of the *** this paper an energy-aware dynamic power management technique based on energy aware earliest deadline first(EA-EDF)scheduling is proposed for improving the performance and reliability by reducing energy and power consumption in the system on chip(SOC).Dynamic power management(DPM)enables MPSOC to reduce power and energy consumption by adopting a suitable core configuration for task *** migration avoids peak temperature values in the multicore *** utilization factor(ui)on central processing unit(CPU)core consumes more energy and increases the temperature *** technique switches the core bymigrating such task to a core that has less temperature and is in a low power *** proposed EA-EDF scheduling technique migrates load on different cores to attain stability in temperature among multiple cores of the CPU and optimized the duration of the idle and sleep periods to enable the low-temperature *** effectiveness of the EA-EDF approach reduces the utilization and energy consumption compared to other existing methods and *** simulation results show the improvement in performance by optimizing 4.8%on u_(i) 9%,16%,23%and 25%at 520 MHz operating frequency as compared to other energy-aware techniques for MPSoCs when the least number of tasks is in running state and can

关键词： Dynamic power management dynamic voltage&frequency scaling dynamic thermal management multiprocessor system on chip complementary metal oxide semiconductor reliability

来源：评论

学校读者我要写书评

暂无评论

Joint fuzzy background and adaptive foreground model for moving target detection

引用

Frontiers of computer science 2024年第2期18卷 93-106页

作者： Dawei ZHANG Peng WANG Yongfeng DONG Linhao LI Xin LI School of Artificial Intelligence Hebei University of TechnologyTianjin 300401China Lane Department of Computer Science and Electrical Engineering West Virginia UniversityMorgantownWV 26506-6109USA

Moving target detection is one of the most basic tasks in computer *** conventional wisdom,the problem is solved by iterative optimization under either Matrix Decomposition(MD)or Matrix Factorization(MF)*** utilizes foreground information to facilitate background *** uses noise-based weights to fine-tune the *** both noise and foreground information contribute to the recovery of the *** jointly exploit their advantages,inspired by two framework complementary characteristics,we propose to simultaneously exploit the advantages of these two optimizing approaches in a unified framework called Joint Matrix Decomposition and Factorization(JMDF).To improve background extraction,a fuzzy factorization is *** fuzzy membership of the background/foreground association is calculated during the factorization process to distinguish their contributions of both to background *** describe the spatio-temporal continuity of foreground more accurately,we propose to incorporate the first order temporal difference into the group sparsity constraint *** temporal constraint is adjusted *** foreground and the background are jointly estimated through an effective alternate optimization process,and the noise can be modeled with the specific probability *** experimental results of vast real videos illustrate the effectiveness of our *** with the current state-of-the-art technology,our method can usually form the clearer background and extract the more accurate ***-noise experiments show the noise robustness of our method.

关键词： matrixdecomposition matrix factorization generalized sparsity noise modeling

来源：评论

学校读者我要写书评

暂无评论

Towards Web Platform for Cloud-Based Virtual Commissioning of IEC 61499 Distributed Automation Systems 29

Towards Web Platform for Cloud-Based Virtual Commissioning o...

引用

29th IEEE International Conference on Emerging Technologies and Factory Automation, ETFA 2024

作者： Lyu, Tuojian Nguyen, Tung Rumiantsev, Roman Kolesnikov, Mikhail Perera, Hansani Atmojo, Udayanto Dwi Vyatkin, Valeriy Aalto University Department of Electrical Engineering and Automation Espoo Finland Luleå Tekniska Universitet Department of Computer Science Computer and Space Engineering Sweden

ISBN: (纸本)9798350361230

In this paper, we propose a cloud-based virtual commissioning Web platform for IEC 61499 compliant distributed automation systems that support Create-Read-Update-Delete operations for SoftPLC containers from various versions and provide a remote engineering workstation with an embedded web Integrated Development Environment (IDE), Human Machine Interface (HMI), and 3D simulation models. A water tank use case demonstrates the platform's flexibility, enabling control software testing entirely via a Web browser without local installations. © 2024 IEEE.

关键词： Cloud platforms

来源：评论

学校读者我要写书评

暂无评论

Multi-Period Resilient Model for VSC-Based AC-DC HDS Considering Public-Safety Power Shutoff to Mitigate Wildfires

引用

CSEE Journal of Power and Energy Systems 2024年第2期10卷 821-833页

作者： Zekai Wang Tao Ding Xiaosheng Zhang Chenggang Mu Pengwei Du Fangxing Li School of Electrical Engineering Xi'an Jiaotong UniversityXi'an 710049China Electric Reliability Council of Texas AustinTX76574 USA Department of Electrical Engineering and Computer Science The University of TennesseeKnoxville(UTK)TN37996USA

This paper proposes a voltage source converter (VSC) -based AC-DC hybrid distribution system (HDS) resilient model to mitigate power outages caused by wildfires. Before a wildfire happens, the public-safety power shutoff (PSPS) strategy is applied to actively cut some vulnerable lines which may easily cause wildfires, and reinforce some lines that are connected to critical loads. To mitigate load shedding caused by active line disconnection in the PSPS strategy, network reconfiguration is applied before the wildfire occurrence. During the restoration period, repair crews (RCs) repair faulted lines, and network reconfiguration is also taken into consideration in the recovery strategy to pick up critical loads. Since there exists possible errors in the wildfire prediction, several different scenarios of wildfire occurrence have been taken into consideration, leading to the proposition of a stochastic multi-period resilient model for the VSC-based AC-DC HDS. To accelerate the computational performance, a progressive hedging algorithm has been applied to solve the stochastic model which can be written as a mixed-integer linear program. The proposed model is verified on a 106-bus AC-DC HDS under wildfire conditions, and the result shows the proposed model not only can improve the system resilience but also accelerate computational speed.

关键词： AC-DC hybrid distribution system restoration network reconfiguration progressive hedging public safety power shutoffs strategy resilience

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：