检索结果-内蒙古大学图书馆

Clustered Federated Multi-Task Learning: A communication-and-Computation Efficient Sparse Sharing Approach

IEEE Transactions on Wireless communications 2025年第6期24卷 4824-4838页

作者： Ai, Yuhan Chen, Qimei Zhu, Guangxu Wen, Dingzhu Jiang, Hao Zeng, Jun Li, Ming Wuhan University School of Electronic Information Wuhan430072 China Shenzhen Research Institute of Big Data Shenzhen518172 China ShanghaiTech University School of Information Science and Technology Shanghai201210 China Fiberhome Communication Technology Company Ltd. Hubei Wuhan430074 China

Federated multi-task learning (FMTL) is a promising technology to tackle one of the most severe non-independent and identically distributed (non-IID) data challenge in federated learning (FL), which treats each client as a single task and learns personalized models by exploiting task correlations. However, the transmission of individual task models generally results in a significant amount of communication overhead compared with global model broadcasting. Furthermore, related works mainly focus on FMTLs with default and static relationships among tasks, which obliterates the non-IID data characteristic. To address these issues, we propose a novel Clustered FMTL mechanism via Sparse Sharing (FedSS). Specifically, we introduce an iterative model pruning approach that trains customized client models to deal with the non-IID issue. Thereafter, we divide clients into different tasks according to their model similarities to promote communication efficiency. Based on clustered tasks, we introduce a sparse sharing mechanism that allows clients to share model parameters dynamically among different tasks to further boost the training performance. On the other aspect, the infertile communication resources would degrade the FMTL performance by restricting the personalized model transmissions. Hence, we first theoretically analyze the convergence performance of the proposed FedSS, which quantitatively unveils the relationship between the local model training performance and communication resources. Thereafter, we formulate a communication-and-computation efficient optimization problem via a joint sparsity ratio assignment and bandwidth allocation strategy. Closed-form expressions for the optimal sparsity ratio and bandwidth allocation are derived based on Lyapunov optimization and block coordinate update (BCU) algorithms. Numerical results illustrate that the proposed FedSS outperforms the benchmarks, and achieves an efficient communication and computation performance. © 2002-20

关键词： Bandwidth

来源：评论

学校读者我要写书评

暂无评论

Utilizing Web Scraping and Machine Learning for Improving Accuracy of Smartphone Price Prediction

Utilizing Web Scraping and Machine Learning for Improving Ac...

引用

2024 International Conference on Advances in data Engineering and Intelligent Computing Systems, ADICS 2024

作者： Vasanth Kumar, V. Tamizhselvan, C. Manoharan, L. Vishnu Sekhar, R. Vinoth Kumar, V. Hindustan Institute Of Technology And Science Department Of Electronics And Communication Engineering Chennai India School Of Computing Srm Institute Of Science And Technology Department Of Data Science And Business Systems Kattankulathur India Hindustan Institute Of Technology And Science Department Of Computer Science And Engineering Chennai India School Of Bioengineering Srm Institute Of Science And Technology Integrated Bioprocessing Laboratory Department Of Biotechnology Kattankulathur India

ISBN: (纸本)9798350364828

Mobile prices play a pivotal role in determining their popularity amongst consumers and their competitive standing within the market. As customers consider their budget while evaluating a mobile phone's specifications and design, estimating the price beforehand becomes essential to cater to their needs and expectations. Hence, accurately forecasting mobile phone prices is a vital step in the product launch process to remain competitive and assess the market dynamics and competitors. The primary aim of this research is to determine the true market value of mobile phones. In this research, the dataset was extracted by means of utilizing a web scraping process from the flipkart website. The main objective of this research is to fetch the most recent dataset based on the most recent features. The extracted dataset is fitted into regression models such as linear regression, decision tree regressor, random forest regressor, KNN regressor, and gradient boosting regressor and performs the comparative analysis. The models are evaluated based on the mean absolute error metric. © 2024 IEEE.

关键词： Decision trees

来源：评论

学校读者我要写书评

暂无评论

Artificial Neural Network Based Image Fusion for Surveillance Application

Artificial Neural Network Based Image Fusion for Surveillan...

引用

International Conference on Machine Vision and Augmented Intelligence, MAI 2022

作者： Paramanandham, Nirmala Sasithradevi, A. Gnana Poovathy, J. Florence Rakshana, V. School of Electronics and Communication Engineering Vellore Institute of Technology Chennai India Ramaiah Institute of Technology Bengaluru India Centre for Advanced Data Science Vellore Institute of Technology Chennai India

ISBN: (纸本)9789819901883

Image fusion is the process of incorporating the information from the similar or dissimilar modality sensor images. The fused image provides enhanced data that may not be obtained by inspecting the sensor outputs distinctly. A novel image fusion technique is proposed in this work for fusing the infrared and visible images. This work proposes an Artificial Neural Network (ANN)-based image fusion technique using Gaussian and Joint Bilateral Filter. Initially the input images are divided into layers using Gaussian smoothness and the Joint Bilateral Filter. The implementation involves the elimination of fine-scale information and extraction of detailed information. Two layers of rules are used for integrating the divided layers and the integrated layers are utilized for reconstructing the image. An enriched final fused image is generated through ANN for surveillance application. The final results assessed using objective metrics such as Edge dependent fusion Quality Index, Entropy, Naturalness Image Quality Evaluator and Structural Similarity Index. This work outperforms quantitatively and achieves better performance with the reduced complexity. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

Visual Computing Unified Application Using Deep Learning and Computer Vision Techniques

International Journal of Interactive Mobile Technologies

引用

International Journal of Interactive Mobile Technologies 2024年第1期18卷 59-74页

作者： Sowmya, B.J. Meeradevi Seema, S. Dayananda, P. Supreeth, S. Shruthi, G. Rohith, S. Department of Artificial Intelligence and Data Science Ramaiah Institute of Technology Karnataka Bengaluru India Department of Artificial Intelligence and Machine Learning Ramaiah Institute of Technology Karnataka Bengaluru India Department of Computer Science and Engineering M S Ramaiah Institute of Technology Karnataka Bengaluru India Department of Information Technology Manipal Institute of Technology Bengaluru Manipal Academy of Higher Education Karnataka Manipal India School of Computer Science and Engineering REVA University Karnataka Bengaluru India Department of Electronics & Communication Engineering Nagarjuna College of Engineering & Technology Karnataka Bengaluru India

Vision Studio aims to utilize a diverse range of modern deep learning and computer vision principles and techniques to provide a broad array of functionalities in image and video processing. Deep learning is a distinct class of machine learning algorithms that utilize multiple layers to gradually extract more advanced features from raw input. This is beneficial when using a matrix as input for pixels in a photo or frames in a video. Computer vision is a field of artificial intelligence that teaches computers to interpret and comprehend the visual domain. The main functions implemented include deepfake creation, digital ageing (de-ageing), image animation, and deepfake detection. Deepfake creation allows users to utilize deep learning methods, particularly autoencoders, to overlay source images onto a target video. This creates a video of the source person imitating or saying things that the target person does. Digital aging utilizes generative adversarial networks (GANs) to digitally simulate the aging process of an individual. Image animation utilizes first-order motion models to create highly realistic animations from a source image and driving video. Deepfake detection is achieved by using advanced and highly efficient convolutional neural networks (CNNs), primarily employing the EfficientNet family of models. © 2024 by the authors of this article. Published under CC-BY.

关键词： Animation

来源：评论

学校读者我要写书评

暂无评论

A novel myocarditis detection combining deep reinforcement learning and an improved differential evolution algorithm

引用

CAAI Transactions on Intelligence technology 2024年第6期9卷 1347-1360页

作者： Jing Yang Touseef Sadiq Jiale Xiong Muhammad Awais Uzair Aslam Bhatti Roohallah Alizadehsani Juan Manuel Gorriz Department of Computer System and Technology Faculty of Computer Science and Information TechnologyUniversiti MalayaKuala LumpurMalaysia Centre for Artificial Intelligence Research(CAIR) Department of Information and Communication TechnologyUniversity of AgderGrimstadNorway Department of Creative Technologies Air UniversityIslamabadPakistan School of Information and Communication Engineering Hainan UniversityHaikouHainanChina Institute for Intelligent Systems Research and Innovation(IISRI) Deakin UniversityWaurn PondsVictoriaAustralia Data Science and Computational Intelligence Institute University of GranadaGranadaSpain

Myocarditis is a serious cardiovascular ailment that can lead to severe consequences if not promptly *** is triggered by viral infections and presents symptoms such as chest pain and heart *** detection is crucial for successful treatment,and cardiac magnetic resonance imaging(CMR)is a valuable tool for identifying this ***,the detection of myocarditis using CMR images can be challenging due to low contrast,variable noise,and the presence of multiple high CMR slices per *** overcome these challenges,the approach proposed incorporates advanced techniques such as convolutional neural networks(CNNs),an improved differential evolution(DE)algorithm for pre-training,and a reinforcement learning(RL)-based model for *** this method presented a significant challenge due to the imbalanced classification of the Z-Alizadeh Sani myocarditis dataset from Omid Hospital in *** address this,the training process is framed as a sequential decision-making process,where the agent receives higher rewards/penalties for correctly/incorrectly classifying the minority/majority ***,the authors suggest an enhanced DE algorithm to initiate the backpropagation(BP)process,overcoming the initialisation sensitivity issue of gradient-based methods like back-propagation during the training *** effectiveness of the proposed model in diagnosing myocarditis is demonstrated through experimental results based on standard performance ***,this method shows promise in expediting the triage of CMR images for automatic screening,facilitating early detection and successful treatment of myocarditis.

关键词： classification differential evolution myocarditis reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

FERMixNet: An Occlusion Robust Facial Expression Recognition Model With Facial Mixing Augmentation and Mid-Level Representation Learning

引用

IEEE Transactions on Affective Computing 2025年第2期16卷 639-654页

作者： Huang, Yansong Peng, Junjie Zhang, Wenqiang Zhao, Tong Chen, Gan Tan, Shuhua Yi, Fen Wang, Lu Shanghai University School of Computer Engineering and Science Shanghai200444 China Shanghai University School of Computer Engineering and Science Shanghai Institute for Advanced Communication and Data Science Shanghai200444 China Fudan University Academy for Engineering and Technology School of Computer Science and Technology Shanghai200437 China YTO Express Company Ltd. National Logistics Engineering Laboratory Shanghai201708 China

Facial expressions can provide a better understanding of people’s mental status and attitudes towards specific things. However, facial occlusion in real world is an unfavorable phenomenon that greatly affects the performance of facial expression recognition models. Recent works addressing the occlusion problem have primarily relied on attention mechanisms or occlusion discarding methods that focus on non-occluded regions of the face. However, these methods have not achieved a good balance between occlusion robustness and model efficiency. In this paper, we propose a simple and efficient model, called FERMixNet, for occluded facial expression recognition. The model incorporates a novel facial mixing augmentation strategy (FERMix) that generates new training samples by simulating real-world facial occlusion and preserving high expression-related semantic information. By co-training the original and newly generated samples, the model’s occlusion robustness is improved without increasing its complexity during inference. Additionally, to further enhance the model’s occlusion robustness, we include mid-level representation learning in the network to learn the discriminative non-occluded local features of the samples with low computational cost. Extensive experiments on four public facial occlusion datasets: Occlusion-RAF-DB, Occlusion-FERPlus and FED-RO show that the proposed model achieves state-of-the-art results which demonstrates the good robustness of our method for occluded facial expression recognition. Meanwhile, the proposed model also achieves state-of-the-art results on the in-the-wild facial expression datasets RAF-DB, AffectNet-8, and AffectNet-7. It proves that the proposed model has good application prospects in real world. © 2010-2012 IEEE.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Artificial intelligence algorithms for object detection and recognition in video and images

引用

Multimedia Tools and Applications 2025年 1-18页

作者： Dakshinamoorthy, Prabakar Rajaram, Gnanajeyaraman garg, Shruti Murugan, Prabhu Manimaran, A. Sundar, Ramesh Department of Data Science and Business System School of Computing SRM Institute of Science and Technology SRM Nagar Kattankulathur Chennai India Department of Computer Science Engineering Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Tamilnadu Chennai India Birla Institute of Technology Mesra Ranchi India Department of ECE Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Chennai602105 India Department of Computer Science Engineering Saveetha School Of EngineeringSaveetha Institute of Medical and Technical Sciences Chennai6021055 India Department of Netwoking and Communication School of Computing SRM Institute of Science and Technology SRM Nagar Kattankulathur Chennai India

The usage of machine learning and deep learning algorithms have necessitated Artificial Intelligence'. AI is aimed at automating things by limiting human interference. It is widely used in IT, healthcare, finance, and agriculture. It is achieved through several deep learning algorithms that reflect the human brain's intelligence. These AI algorithms can be manipulated according to changing needs and improved efficiency. This paper tries to utilize the developments made in AI technology to classify the images and recognize the objects present in them. One widely used AI algorithm is CNN (Convolutional Neural Networks). The CNN is a deep learning-based algorithm that consists of various layers that extract and filters the parameters present in the images. Some additional layers of ResNet50 and the CNN algorithm are used to extract the parameters to improve image recognition accuracy. The image dataset taken for training and testing the proposed model is imageNet. The images are initially processed before sending them to the proposed model. The proposed model is trained, validated, and tested through the images obtained after the initial processing. The same process is repeated several times until getting the maximum accuracy. The accuracy of the proposed model in terms of image recognition is recorded. The obtained results are compared with other image classification algorithms like VGG16 and VGG19. It is concluded that the proposed model outperforms other traditional methods in terms of accuracy. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Bus priority lane in Bengaluru: A study on its effectiveness and driver stress

引用

Transport Policy 2023年 139卷 39-62页

作者： Karthik, P.N. Rathod, Nihesh Yasodharan, Sarath Lobo, Wilson Sahadevan, Ajeesh Sundaresan, Rajesh Verma, Pratik Institute of Data Science National University of Singapore 117583 Singapore Department of Electrical Communication Engineering Indian Institute of Science Karnataka Bengaluru560012 India Karnataka Bengaluru India Netradyne Technology India Pvt Ltd Karnataka Bengaluru India Department of Electrical Communication Engineering and Robert Bosch Center for Cyber-Physical Systems Indian Institute of Science Karnataka Bengaluru560012 India

This paper studies the effectiveness of the bus priority lane (BPL) for public transport buses in the city of Bengaluru in India. We use the travel times on the BPL corridor as a measure of the effectiveness of the BPL. We find that there is a significant improvement in the travel times after the introduction of the BPL;for the worst 10% of the travel times, we find an improvement between 4% and 28%. Our methodology involves extracting trips on the BPL and computing the travel times for these trips from a time series of GPS information. Our methodology is scalable and can be used to compute the travel times between any two given points in other similar studies. We supplement our results with a novel test (called the D-test) for comparing the levels of stressful driving in the following scenarios: (a) morning peak hours (IST 07:00 h to 11:00 h) versus evening peak hours (IST 17:00 h to 21:00 h), and (b) northward trips versus southward trips on the BPL. We are able to infer that the drivers are generally more stressed during the morning peak hours and during the southward trips on the BPL. Partitioning the BPL into segments, we show that a majority of the segments exhibit similar effectiveness and driver stress trends as the full BPL stretch. We anticipate that corrective measures for the betterment of travel times and driver stress levels (e.g., introducing additional buses subject to vehicle re-balancing constraints, carefully planning the bus schedules to regulate bus traffic throughout the day, etc.) in some segments can lead to further improvements in travel times and reduction in driver stress levels. © 2023 Elsevier Ltd

关键词： Travel time

来源：评论

学校读者我要写书评

暂无评论

A federated learning scheme meets dynamic differential privacy

引用

CAAI Transactions on Intelligence technology 2023年第3期8卷 1087-1100页

作者： Shengnan Guo Xibin Wang Shigong Long Hai Liu Liu Hai Toong Hai Sam State Key Laboratory of Public Big Data College of Computer Science and TechnologyGuizhou UniversityGuiyangChina School of Big Data Key Laboratory of Electric Power Big Data of Guizhou ProvinceGuizhou Institute of TechnologyGuiyangChina School of Information Guizhou University of Finance and EconomicsGuiyangChina Faculty of Business and Communication INTI International UniversityNilaiMalaysia

Federated learning is a widely used distributed learning approach in recent years,however,despite model training from collecting data become to gathering parameters,privacy violations may occur when publishing and sharing models.A dynamic approach is pro-posed to add Gaussian noise more effectively and apply differential privacy to federal deep ***,it is abandoning the traditional way of equally distributing the privacy budget e and adjusting the privacy budget to accommodate gradient descent federation learning dynamically,where the parameters depend on computation derived to avoid the impact on the algorithm that hyperparameters are created *** also incorporates adaptive threshold cropping to control the sensitivity,and finally,moments accountant is used to counting the∈consumed on the privacy‐preserving,and learning is stopped only if the∈_(total)by clients setting is reached,this allows the privacy budget to be adequately explored for model *** experimental results on real datasets show that the method training has almost the same effect as the model learning of non‐privacy,which is significantly better than the differential privacy method used by TensorFlow.

关键词： data privacy machine learning security of data

来源：评论

学校读者我要写书评

暂无评论

Guardians of Truth: Exploring Techniques for Veracity Identification in social media during an Infodemic 3

Guardians of Truth: Exploring Techniques for Veracity Identi...

引用

3rd International Conference on Artificial Intelligence For Internet of Things, AIIoT 2024

作者： Suthanthiradevi, P. Revathy, G. Sathish Kumar, P. Karthikeyan, B. Muthu, Lakshmi.V SRM institute of Science and Technology Data Science and Business Systems School of Computing India Vels Institute of science Technology and Advanced studies Department of Computer science and engineering India Rajalakshmi institute of technology Computer and Communication Engineering India SASTRA Deemed University School of Computing India St. Joseph's Institute of Technology Department of Computer Science and Engineering India

ISBN: (纸本)9798350372120

With the recent advent of technology, social networks are accessible 24/7 using mobile devices. During covid-19 pandemic the propagation of misinformation are mostly related to the disease, its cures and prevention. With Twitter as the medium this information was spreaded to millions of people who lack domain specific knowledge. This can lead to bloster fear and direct damage to the people. This research focuses on identifying the veracity of the Twitter posts behind pandemic situation. The authors have proposed a Transformer Model using Bat Algorithm for identifying the fake news. This model is trained and evaluated using infodemic metadata. Our experimental results shows by which proposed model accurately detected the fake news and achieved 96.5% of f1-score. © 2024 IEEE.

关键词： Social networking (online)

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：