检索结果-内蒙古大学图书馆

FedCau: A Proactive Stop Policy for Communication and Computation Efficient Federated learning

IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS 2024年第9期23卷 11076-11093页

作者： Mahmoudi, Afsaneh Ghadikolaei, Hossein S. Da Silva Jr, Jose Mairton Barros Fischione, Carlo KTH Royal Inst Technol Sch Elect Engn & Comp Sci S-10044 Stockholm Sweden Ericsson S-16483 Stockholm Sweden Uppsala Univ Dept Informat Technol S-75105 Uppsala Sweden

This paper investigates efficient distributed training of a Federated learning (FL) model over a wireless network of wireless devices. The communication iterations of the distributed training algorithm may be substantially deteriorated or even blocked by the effects of the devices' background traffic, packet losses, congestion, or latency. We abstract the communication-computation impacts as an 'iteration cost' and propose a cost-aware causal FL algorithm (FedCau) to tackle this problem. We propose an iteration-termination method that trade-offs the training performance and networking costs. We apply our approach when workers use the slotted-ALOHA, carrier-sense multiple access with collision avoidance (CSMA/CA), and orthogonal frequency-division multiple access (OFDMA) protocols. We show that, given a total cost budget, the training performance degrades as either the background communication traffic or the dimension of the training problem increases. Our results demonstrate the importance of proactively designing optimal cost-efficient stopping criteria to avoid unnecessary communication-computation costs to achieve a marginal FL training improvement. We validate our method by training and testing FL over the MNIST and CIFAR-10 dataset. Finally, we apply our approach to existing communication efficient FL methods from the literature, achieving further efficiency. We conclude that cost-efficient stopping criteria are essential for the success of practical FL over wireless networks.

关键词： Costs Training Wireless networks Protocols Optimization machine learning algorithms Resource management Federated learning communication protocols cost-efficient algorithm latency unfolding federated learning

来源：评论

学校读者我要写书评

暂无评论

Integrating machine learning for enhanced wildfire severity prediction: A study in the Upper Colorado River basin

引用

SCIENCE OF THE TOTAL ENVIRONMENT 2024年 952卷 175914页

作者： Han, Heechan Abitew, Tadesse A. Bazrkar, Hadi Park, Seonggyu Jeong, Jaehak Chosun Univ Dept Civil Engn Gwangju South Korea Texas A&M AgriLife Res Temple TX 76502 USA

Wildfires pose significant threats worldwide, requiring accurate prediction for mitigation. This study uses machine learning techniques to forecast wildfire severity in the Upper Colorado River basin. Datasets from 1984 to 2019 and key indicators like weather conditions and land use were employed. Random Forest outperformed Artificial Neural Network, achieving 72 % accuracy. Influential predictors include air temperature, vapor pressure deficit, NDVI, and fuel moisture. Solar radiation, SPEI, precipitation, and evapotranspiration also contribute significantly. Validation against actual severities from 2016 to 2019 showed mean prediction errors of 11.2 %, affirming the model's reliability. These results highlight the efficacy of machine learning in understanding wildfire severity, especially in vulnerable regions.

关键词： Artificial neural networks Landsat machine learning algorithms Normalized Burn Ratio Random Forest Upper Colorado River basin Wildfire severity prediction

来源：评论

学校读者我要写书评

暂无评论

Assessing the generalization of forecasting ability of machine learning and probabilistic models for complex climate characteristics

引用

STOCHASTIC ENVIRONMENTAL RESEARCH AND RISK ASSESSMENT 2024年第8期38卷 2927-2947页

作者： Batool, Aamina Ali, Zulfiqar Mohsin, Muhammad Masmoudi, Atef Kartal, Veysi Satti, Samina Univ Punjab Coll Stat Sci Lahore Pakistan King Khalid Univ Coll Comp Sci Abha Saudi Arabia Siirt Univ Engn Fac Civil Engn Siirt Turkiye Univ Wah Dept Stat Rawalpindi Pakistan

Climate changes and global warming increase risk of recurrent extreme and complex climatic features. It necessitates accurate modeling and forecasting of climate phenomena for sustainable development goals. However, machine learning algorithms and advanced statistical models are extensively employed to analyze complex data and make predictions related to climate phenomena. It is important to have comprehensive knowledge to use these models and consider their potential implications. This study aims to evaluate and compare some popular machine learning and probabilistic methods by analyzing various time series indices associated with precipitation and temperature. For application, time series data of Standardized Precipitation Temperature Index (SPTI), Standardized Temperature Index (STI), Standardized Compound Drought and Heat Index (SCDHI), and Biased Diminished Weighted Regional Drought Index (BDWRDI) are used from various meteorological regions of Pakistan. The performance of each algorithm is compared using Residual Mean Square Error (RMSE) and Mean Average Error (MAE). The outcomes associated with this research indicate a higher preference of neural networks over machine learning methods in the training sets. However, the efficiency varies from model to model, indicator to indicator, time scale to time scale, and location to location during the testing phase. The most appropriate models are found by considering a list of candidates forecasting models and investigating the performance of each model.

关键词： Climate change machine learning algorithms Advanced statistical models Efficiency

来源：评论

学校读者我要写书评

暂无评论

Adapting to the Evolution: Enhancing Intrusion Detection Through machine learning in the QUIC Protocol Era

引用

IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT 2025年第2期22卷 1929-1944页

作者： Kadi, Adam Khoukhi, Lyes Viinikka, Jouni Fabre, Pierre-Edouard Normandie Univ ENSICAEN UNICAEN CNRSGREYC F-14000 Caen France 6cure F-14200 Caen France

The advent of the QUIC protocol may herald a significant shift in the composition of online traffic in the years to come. The transport layer encryption of the QUIC protocol is one of its main evolutions, especially for metadata that was previously transmitted over TCP traffic without encryption. This new protocol has the potential to require significant alterations in future Internet traffic analysis methods and impact network intrusion detection. On the other side, machine learning has been used in several research projects to identify network intrusions, with positive outcomes. However, we must take into account new evolution of network traffic. In this paper, we propose a new approach that employs supervised machine learning algorithms to identify flows generated by bots interacting with a Web server during a DDoS attack, focusing on the challenges posed by the QUIC protocol and its implications for effective intrusion detection and cybersecurity. Our contribution in this work is divided into three main parts: 1) A guided process with model architecture for emulating and collecting traffic that depict a range of situations our system may encounter;2) an analysis module that consists on the creation of two labeled datasets, where observations represent the traffic flows detected in PCAP files. We studied the relevance of different features for these datasets, contributing to a thorough understanding of the quality of the data used;3) a real world experimention for evaluating the effectiveness of several supervised machine learning algorithms on our datasets. This experimentation allows us to determine which algorithm provides the best prediction results.

关键词： Denial-of-service attack Protocols machine learning algorithms machine learning Floods Computer crime Botnet Accuracy Mathematical models Intrusion detection DDoS QUIC HTTP/3 cybersecurity intrusion detection system

来源：评论

学校读者我要写书评

暂无评论

Auto-Updating Intrusion Detection System for Vehicular Network: A Deep learning Approach Based on Cloud-Edge-Vehicle Collaboration

引用

IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY 2024年第10期73卷 15372-15384页

作者： Fan, Chunyang Cui, Jie Jin, Hulin Zhong, Hong Bolodurina, Irina He, Debiao Anhui Univ Sch Comp Sci & Technol Key Lab Intelligent Comp & Signal Proc Minist Educ Hefei 230039 Peoples R China Anhui Univ Anhui Engn Lab IoT Secur Technol Hefei 230039 Peoples R China Orenburg State Univ Fac Math & Informat Technol Orenburg 460018 Russia Wuhan Univ Sch Cyber Sci & Engn Wuhan 430072 Peoples R China Matrix Elements Technol Shanghai Key Lab Privacy Preserving Computat Shanghai 201204 Peoples R China

Intrusion detection systems play a crucial role in ensuring the safety of vehicle driving. Traditional intrusion detection systems face challenges in efficiently extracting key features from network traffic, resulting in inefficiency. To obtain better detection performance, many researchers adopted deep learning to build intrusion detection systems for vehicular network. However, these systems can only identify learned network attacks, while failing to identify the unknown attacks present in open scenarios. In this paper, we propose a deep learning-based auto-updating intrusion detection system for vehicular network, which uses a two-layer filter to effectively sense unknown attacks in open scenarios and proactively updates the attack recognizer to enable it with the ability to identify unknown attacks. In addition, we design a cloud-edge-vehicle collaborative model updating approach, where the updating task of the attack recognizer is deployed on a cloud server to reduce the computational load of the vehicle. The comprehensive experiments on three open scenarios show that the updated attack recognizer can accurately identify both known and unknown attacks.

关键词： Intrusion detection Feature extraction Telecommunication traffic Deep learning Servers machine learning algorithms Computational modeling Vehicular network intrusion detection system deep learning unknown attack

来源：评论

学校读者我要写书评

暂无评论

A survey on machine learning algorithm applications in visible light communication systems

引用

OPTICAL AND QUANTUM ELECTRONICS 2024年第8期56卷 1-23页

作者： Sliti, Maha Mrabet, Manel Garai, Mouna Ammar, Lassaad Ben Univ Carthage Higher Sch Commun Tunis SUPCOM Commun Networks & Secur Res Lab LR11TIC04 Tunis Tunisia Univ Carthage Higher Sch Commun Tunis SUPCOM Green & Smart Commun Syst Res Lab LR11TIC02 Tunis Tunisia Prince Sattam Bin Abdulaziz Univ Al Kharj Saudi Arabia Manouba Univ Manouba Tunisia Higher Inst Technol Studies Med Medenine Tunisia

Visible Light Communication (VLC) technology offers a promising alternative for high-speed data transfer due to its ability to simultaneously illuminate and transport data. This research studies the application of machine learning (ML) algorithms for the purpose of addressing several issues associated with VLC technology. These challenges include the reduction of non-linearity, channel estimation, detection of modulation format, localization, enhancement of security, optimization of mobility, and management of resources. ML algorithms have the potential to enhance the efficiency and reliability of VLC systems, ensuring that data flow is uninterrupted. Additionally, the study analyzes the challenges that are involved with the application of ML algorithms in VLC technology as well as the potential future research topics that could be pursued.

关键词： Visible light communication (VLC) machine learning algorithms Applications Open research directions

来源：评论

学校读者我要写书评

暂无评论

Achieving Equity via Transfer learning With Fairness Optimization

引用

IEEE ACCESS 2024年 12卷 195229-195241页

作者： Wang, Xiaoyang Chang, Chia-Hsuan Yang, Christopher C. Drexel Univ Coll Comp & Informat Philadelphia PA 19104 USA

machine learning algorithms are increasingly used in real-world decision-making systems, raising concerns about potential biases and unfairness. Existing in-processing bias mitigation approaches often focus on achieving numerical parity across demographic groups while neglecting the performance impact on individual groups and the overall performance-fairness trade-off. This can lead to increased discriminatory outcomes and hinder efforts to develop truly fair AI systems. This paper proposes Transfer learning with Fairness Optimization (TLFO), a novel framework that serially optimizes predictive performance and fairness in machine learning models. TLFO leverages transfer learning by dividing the training process into two distinct phases: (1) initial learning for performance optimization and (2) subsequent fine-tuning for fairness enhancement. This sequential approach enables fine-grained control over fairness constraints, minimizing the performance-fairness trade-off. Extensive experiments on two real-world datasets demonstrate TLFO's effectiveness. TLFO consistently achieves superior fairness with minimal performance degradation compared to state-of-the-art in-processing bias mitigation approaches, highlighting its potential for generating fair and accurate classifiers with versatile applications.

关键词： Optimization Transfer learning Training Predictive models machine learning algorithms Prevention and mitigation Adaptation models Data models Accuracy Transforms AI fairness transfer learning in-processing bias mitigation performance-fairness trade-off

来源：评论

学校读者我要写书评

暂无评论

Novel Statistical Regularized Extreme learning Algorithm to Address the Multicollinearity in machine learning

引用

IEEE ACCESS 2024年 12卷 102355-102367页

作者： Yildirim, Hasan Karamanoglu Mehmetbey Univ Kamil Ozdag Fac Sci Dept Math TR-70100 Karaman Turkiye

The multicollinearity problem is a common phenomenon in data-driven studies, significantly affecting the performance of machine learning algorithms during the process of extracting information from data. Despite its widespread use across various fields, the extreme learning machine (ELM) also suffers from multicollinearity issues. To address this challenge, the ridge and Liu estimators, drawn from statistics literature, have been integrated into ELM theory, resulting in a notable advancement. This study aims to further enhance the capabilities of ridge and Liu estimators within the ELM framework by introducing two innovative two-parameter algorithms (TP1-ELM and TP2-ELM) that simultaneously incorporate both estimators. The proposed algorithms undergo comprehensive benchmarking against ELM, ELM-based algorithms, and other commonly used machine learning techniques across seven diverse datasets. Benchmark results demonstrate that the proposed algorithms consistently outperform both ELM-focused approaches and traditional machine learning algorithms on most datasets, yielding more generalizable and stable results. These findings suggest that the proposed algorithms offer a promising alternative to traditional machine learning techniques for regression and classification tasks, particularly in scenarios where multicollinearity is a concern.

关键词： machine learning algorithms Classification algorithms Tuning machine learning Prediction algorithms Extreme learning machines Computational modeling Extreme learning machine Liu estimator machine learning multicollinearity ridge estimator Tikhonov regularization

来源：评论

学校读者我要写书评

暂无评论

A Scalable and Adaptable Supervised learning Approach for Solving the Traveling Salesman Problems

引用

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS 2024年第11期25卷 17092-17104页

作者： Lyu, Zefeng Islam, Md. Zahidul Yu, Andrew Junfang Univ Tennessee Knoxville Dept Ind & Syst Engn Knoxville TN 37996 USA Clemson Univ Dept Ind Engn Clemson SC 29634 USA

The Traveling Salesman Problem (TSP) is a well-known combinatorial optimization problem that has attracted extensive research efforts in developing exact methods and heuristics. However, the scalability and generalization of learning-based approaches remain significant challenges. This paper addresses this gap by proposing an approach that integrates supervised learning with traditional approaches. In particular, we introduce the concept of "anchors", which represents nodes that should be connected to their nearest neighbors in the optimal solution. Our approach differs from the previous supervised learning approaches in that, instead of using the whole distance matrix, our neural network leverages local information to make predictions, which enables it to handle arbitrarily large-scale TSP instances without a decline on prediction accuracy. Experimental results demonstrate that our model successfully identifies 87% of the anchors with a precision of over 95% for both generated instances and TSPLIB instances, which were unseen by the model during the training. The performance of the proposed framework is evaluated by integrating the predicted anchors into existing methods such as the Miller-Tucker-Zemlin (MTZ) model and the insertion algorithms. Our approach demonstrates remarkable improvements in solution quality and computational time. By comparing with the graph pointer network, a state-of-the-art learning-based approach, the proposed algorithm achieves a 28% reduction in the average gap and a 59% decrease in computational time for solving TSP instances with 1,000 nodes.

关键词： Supervised learning Prediction algorithms machine learning algorithms Scalability Heuristic algorithms Training Computational modeling Traveling salesman problem supervised learning constructive heuristics integer linear programming generalization adaptability

来源：评论

学校读者我要写书评

暂无评论

Defect analysis of 3D printed object using transfer learning approaches

引用

EXPERT SYSTEMS WITH APPLICATIONS 2024年 253卷

作者： Ahsan, Md Manjurul Raman, Shivakumar Liu, Yingtao Siddique, Zahed Univ Oklahoma Sch Ind & Syst Engn Norman OK 73019 USA Univ Oklahoma Sch Aerosp & Mech Engn Norman OK 73019 USA

Additive manufacturing (AM) is rapidly gaining traction across diverse industries, including healthcare, aerospace, and automotive, mainly due to its ability to enhance production efficiency. However, a critical challenge in AM is the early detection of defects, which can significantly reduce production costs and improve productivity. Addressing this need, our study explored the effectiveness of machine learning (ML) approaches, mainly focusing on transfer learning (TL) models, for defect detection in 3D-printed objects. We utilized a range of TL models such as visualgeometrygroup (VGG)-16, inception-residualnetwork (InceptionResNet)V2, residual network (ResNet)-50, mobilenetwork (MobileNet)-V2, VGG19, ResNet101, extreme inception (Xception), efficientnetwork (EfficientNet)-B0, EfficientNetB7, efficientnetworkV2medium (EfficientNetV2M), and neural architecture search network large (NASNetLarge), analyzing images of 3D-printed objects across three datasets using various statistical measures. An ablation study was also conducted using adaptive moment estimation (Adam), stochastic gradient descent (Sgd), and root mean square propagation (Rmsprop) optimizers. We identified that TL models such as VGG16, MobileNetV2, InceptionResNetV2, and NASNetLarge significantly outperform others, especially when optimized with Adam and Rmsprop. Conversely, we found that models like EfficientNetB0, EfficientNetB7, and EfficientNetV2M exhibited lower performance when paired with the Sgd optimizer. Additionally, Local Interpretable Model-agnostic (LIME)-based approaches are used to provide explanations of the models' predictions. These findings offer substantial insights into model optimization and integration, aiming to enhance AM product quality through advanced image-based monitoring and inspection.

关键词： Additive manufacturing Defect detection Transfer learning 3D printed objects machine learning algorithms LIME

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：