检索结果-内蒙古大学图书馆

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Yi, Liping Yu, Han Ren, Chao Wang, Gang Liu, Xiaoguang Li, Xiaoxiao College of Computer Science TMCC SysNet DISSec GTIISC Nankai University China College of Computing and Data Science Nanyang Technological University Singapore Department of Electrical and Computer Engineering The University of British Columbia Canada Vector Institute Canada

Model heterogeneous federated learning (MHeteroFL) enables FL clients to collaboratively train models with heterogeneous structures in a distributed fashion. However, existing MHeteroFL methods rely on training loss to transfer knowledge between the client model and the server model, resulting in limited knowledge exchange. To address this limitation, we propose the Federated model heterogeneous Matryoshka Representation Learning (FedMRL) approach for supervised learning tasks. It adds an auxiliary small homogeneous model shared by clients with heterogeneous local models. (1) The generalized and personalized representations extracted by the two models' feature extractors are fused by a personalized lightweight representation projector. This step enables representation fusion to adapt to local data distribution. (2) The fused representation is then used to construct Matryoshka representations with multi-dimensional and multi-granular embedded representations learned by the global homogeneous model header and the local heterogeneous model header. This step facilitates multi-perspective representation learning and improves model learning capability. Theoretical analysis shows that FedMRL achieves a O(1/T) non-convex convergence rate. Extensive experiments on benchmark datasets demonstrate its superior model accuracy with low communication and computational costs compared to seven state-of-the-art baselines. It achieves up to 8.48% and 24.94% accuracy improvement compared with the state-of-the-art and the best same-category baseline, respectively. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Deep Learning-Cryptography Hybrid Approach for Ensuring Medical Image Confidentiality

A Deep Learning-Cryptography Hybrid Approach for Ensuring Me...

引用

Electrical, Electronics, Information and Communication Technologies (ICEEICT), International Conference on

作者： Hitesh Surya Chinta Maganti Akshita Uma Sai Kavitha C.R Department of Computer Science & Engineering Amrita School of Computing Bengaluru India

ISBN: (数字)9798350369083

ISBN: (纸本)9798350369090

This paper addresses the important need for security measures in the transmission of medical images to ensure the confidentiality of sensitive patient information. Leveraging a synergistic combination of advanced cryptography and steganography techniques, including DNA-based AES encryption and LSB embedding, the proposed system ensures the secure hiding and transmission of patient data. Furthermore, Discrete Wavelet Transform compression is employed to optimize data size while preserving integrity and also applied deep learning models to the HAM10000 dataset. Experimental evaluations prove the superiority and efficacy of the suggested strategy in safeguarding medical data against unauthorized access and ensuring confidentiality. This work contributes significantly to the advancement of secure medical data transmission systems, addressing security concerns and regulatory requirements in the healthcare domain.

关键词： Deep learning Steganography Image coding Transforms Medical services Minimization Discrete wavelet transforms Encryption Protection Biomedical imaging

来源：评论

学校读者我要写书评

暂无评论

Comparative study of IoT-and AI-based computing disease detection approaches

引用

Data science and Management 2025年第1期8卷 94-106页

作者： Wasiur Rhmann Jalaluddin Khan Ghufran Ahmad Khan Zubair Ashraf Babita Pandey Mohammad Ahmar Khan Ashraf Ali Amaan Ishrat Abdulrahman Abdullah Alghamdi Bilal Ahamad Mohammad Khaja Shaik Department of Computer Application Lovely Professional UniversityPhagwaraPunjab144411India Department of Computer Science and Engineering Koneru Lakshmaiah Education FoundationGunturAndhra Pradesh522502India School of Computing Science and Engineering VIT Bhopal UniversityKothrikalanSehore466114India Babasaheb Bhimrao Ambedkar University Lucknow226025India College of Commerce and Business Administration Dhofar UniversitySalalah211Oman Faculty of Computer Studies Arab Open University18211Bahrain Department of Computer Application Shri Ramswaroop Memorial UniversityLucknow225003India College of Computing and Information Technology Shaqra UniversityShaqra11961Saudi Arabia

The emergence of different computing methods such as cloud-,fog-,and edge-based Internet of Things(IoT)systems has provided the opportunity to develop intelligent systems for disease *** to other machine learning models,deep learning models have gained more attention from the research community,as they have shown better results with a large volume of data compared to shallow ***,no comprehensive survey has been conducted on integrated IoT-and computing-based systems that deploy deep learning for disease *** study evaluated different machine learning and deep learning algorithms and their hybrid and optimized algorithms for IoT-based disease detection,using the most recent papers on IoT-based disease detection systems that include computing approaches,such as cloud,edge,and *** analysis focused on an IoT deep learning architecture suitable for disease *** also recognizes the different factors that require the attention of researchers to develop better IoT disease detection *** study can be helpful to researchers interested in developing better IoT-based disease detection and prediction systems based on deep learning using hybrid algorithms.

关键词： Deep learning Internet of Things(IoT) Cloud computing Fog computing Edge computing

来源：评论

学校读者我要写书评

暂无评论

AI-Driven Insights: A Survey on Innovative Approach for Lung Cancer Prediction Utilizing Machine Learning and Deep Learning Methods : Lung Cancer Prediction Utilizing Machine Learning and Deep Learning Methods

AI-Driven Insights: A Survey on Innovative Approach for Lung...

引用

Automation, computing and Renewable Systems (ICACRS), International Conference on

作者： A. Lakshmi Bhargav C. Ashokkumar Department of Computing Technologies School of Computing College of Engineering and Technology SRM Institute of Science & Technology Chennai India

ISBN: (数字)9798331532420

ISBN: (纸本)9798331532437

The lungs are vital for exchanging oxygen and carbon dioxide, providing oxygen to cells for energy production, and removing waste carbon dioxide from the *** also help filter and protect the body from harmful particles and pathogens in the *** lung function is essential for overall health and *** cancer ranks as the second most common cause of cancer-related fatalities *** emphasizes the challenge of early detection due to the lack of symptoms in the initial *** contributes to a high mortality rate compared to other lung *** identification and precise prediction of lung cancer serves as vital for improving patient outcomes. Machine learning (ML) techniques have shown great promise in revolutionizing lung cancer prediction by leveraging various data sources such as medical images, clinical data, and genomic information. This paper presents a review of lung cancer prediction using ML, DL, and AI *** classification of lung cancer through medical imaging is essential for physicians to determine appropriate treatment options, ultimately aiming to reduce cancer *** goal is to compare the performance and accuracy of convolutional neural networks (CNNs), Random forest algorithm, ensemble extreme boosting (XGBoost) algorithm,Support Vector Machine (SVM), AdaBoost classification model (ADB-C), Long Short-Term Memory Networks (LSTM), incremental multiple resolution residual network (iMRRN) models while also providing explanations for their predictions.

关键词： Deep learning Machine learning algorithms Lungs Lung cancer Carbon dioxide Predictive models Prediction algorithms Classification algorithms Prognostics and health management Long short term memory

来源：评论

学校读者我要写书评

暂无评论

Energy Consumption Forecasting in Buildings based on heating and cooling loads using Regression models 4

Energy Consumption Forecasting in Buildings based on heating...

引用

4th IEEE International Conference on Smart Technologies in computing, Electrical and Electronics, ICSTCEE 2023

作者： Samhitha, D. Sai Arjun, D. Shashank Padmavathi, A. Hemprasanna, A. Amrita Vishwa Vidyapeetham Amrita School Of Computing Department Of Computer Science And Engineering Chennai India Amrita Vishwa Vidyapeetham Amrita School Of Engineering Department Of Electrical And Computer Coimbatore India

ISBN: (纸本)9798350384628

The Energy consumption in buildings is a significant factor in both cost-effectiveness and environmental sustainability. To optimize energy use and sustain heating, cooling, and air conditioning (HVAC) in the buildings, predictive models that can effectively estimate heating and cooling loads are crucial. This paper employs a variety of regression techniques to develop prediction models for predicting Cooling Loads (CL) Heating Loads (HL) and in buildings. XG Boost, Ridge regression, Lasso regression, Decision Tree, Random Forest, and Bayesian Ridge Regression are a few ML models used. Numerous building metrics that are connected to energy are included in the collection. The models are trained, tested, and optimized using metrics like R2 score, MAE-Mean Absolute Error, MSE-Mean Squared Error, and RMSE-Root Mean Squared Error. To improve its capacity for prediction, the model's hyperparameters are also created using GridSearchCV. Analysis of the advantages and disadvantages of each model is feasible thanks to the results' systematic comparison and visualization. The results underline how crucial it is to choose models carefully and accurately to increase energy efficiency and inform building designs. This study offers useful tips on how to use ensemble techniques and machine learning techniques to evaluate power usage in the buildings. © 2023 IEEE.

关键词： Energy utilization

来源：评论

学校读者我要写书评

暂无评论

Identification of Key Signaling Pathways and Novel Computational Drug Target for Depression and Coronary Artery Disease

Identification of Key Signaling Pathways and Novel Computati...

引用

2024 IEEE International Conference on computing, Applications and Systems, COMPAS 2024

作者： Hosen, Md. Faruk Basar, Md. Abul Yasmin, Mst. Farjana Morshed, Monir Uddin, Muhammad Shahin Daffodil International University Department of Computing and Information System Dhaka1216 Bangladesh Department of Computer Science and Engineering Dhaka1230 Bangladesh Mawlana Bhashani Science and Technology University Department of Information and Communication Technology Tangail1902 Bangladesh

ISBN: (纸本)9798331529765

Psychological disorders, such as anxiety, bipolar disorder, panic disorder, stress, depression, and schizophrenia, are increasingly prevalent worldwide. Among these conditions, depression is particularly notable as one of the most common and debilitating neuropsychiatric disorders. Individuals with depression may be at a higher risk of developing oronary Artery Disease (CAD). Depression can contribute to poor lifestyle choices, such as unhealthy eating, lack of exercise, and smoking, which are risk factors for CAD. The emotional stress and anxiety associated with depression can strain the heart and exacerbate CAD symptoms. The relationship is not one-sided. CAD itself can be a significant source of emotional distress, leading to symptoms of depression and anxiety in affected individuals. Managing both conditions in tandem can be complicated. Treating CAD may involve medications, lifestyle modifications, and potentially surgical interventions. Meanwhile, depression often requires therapy, counseling, and medication. Coordinating care and addressing both conditions simultaneously is crucial. In our study, we investigated the molecular connections between CAD and Depression using GSE98793 and GSE20681 microarray datasets. After preprocessing, we identified key hub genes, including CCT2, SVIL, REPS2, ASPH, and UBC, in the shared ProteinProtein Interaction network. KEGG pathways linked these DEGs to colorectal and cancer pathways. Our next steps involve exploring microRNAs, TFs, and GO analysis. These findings offer promising leads for potential therapies, uniting CAD and Depression under a common molecular framework, advancing our understanding of these conditions. © 2024 IEEE.

关键词： Blood vessels

来源：评论

学校读者我要写书评

暂无评论

A Creative Research in the Field of Urban Fire Detection 5

A Creative Research in the Field of Urban Fire Detection

引用

5th International Conference on Computer Vision, Image and Deep Learning, CVIDL 2024

作者： Feng, Junming Wang, Ruixiang Xi, Chenyang Peng, Jiayu Harbin Institute of Technology Faculty of Computing Harbin China Harbin Institute of Technology School of Information Science and Engineering Harbin China Harbin Institute of Technology School of Software Harbin China

ISBN: (纸本)9798350373820

As the most common disaster in human daily production and life, fire is very likely to threaten people's life and property safety. In order to avoid fire, this paper designs and implements a fire detection system based on video and YOLOv5 model, and describes the improvement of YOLOv5 compared with the previous generation YOLO model. Collect fire images through the network, use image processing technology to enhance the data, make fire data set and divide the data set by setting aside method. Then the self-made fire data set is used to train the fire detection model based on YOLOv5 model, and the model is improved by adding transfer learning, multi-scale training and other technologies to improve the fire detection rate, and the corresponding weight files are obtained. Finally, the training method of the model and the ability of the model to detect fire are expounded, and the innovative application of the model in the field of fire detection is realized. © 2024 IEEE.

关键词： Fires

来源：评论

学校读者我要写书评

暂无评论

Self-training through Classifier Disagreement for Cross-Domain Opinion Target Extraction 23

Self-training through Classifier Disagreement for Cross-Doma...

引用

32nd ACM World Wide Web Conference, WWW 2023

作者： Sun, Kai Zhang, Richong Samuel, Mensah Nikolaos, Aletras Mao, Yongyi Liu, Xudong Beijing Advanced Innovation Center for Big Data and Brain Computing School of Computer Science and Engineering Beihang University Beijing China Department of Computer Science University of Sheffield Sheffield United Kingdom School of Electrical Engineering and Computer Science University of Ottawa Ottawa Canada

ISBN: (纸本)9781450394161

Opinion target extraction (OTE) or aspect extraction (AE) is a fundamental task in opinion mining that aims to extract the targets (or aspects) on which opinions have been expressed. Recent work focus on cross-domain OTE, which is typically encountered in real-world scenarios, where the testing and training distributions differ. Most methods use domain adversarial neural networks that aim to reduce the domain gap between the labelled source and unlabelled target domains to improve target domain performance. However, this approach only aligns feature distributions and does not account for class-wise feature alignment, leading to suboptimal results. Semi-supervised learning (SSL) has been explored as a solution, but is limited by the quality of pseudo-labels generated by the model. Inspired by the theoretical foundations in domain adaptation [2], we propose a new SSL approach that opts for selecting target samples whose model output from a domain-specific teacher and student network disagree on the unlabelled target data, in an effort to boost the target domain performance. Extensive experiments on benchmark cross-domain OTE datasets show that this approach is effective and performs consistently well in settings with large domain shifts. © 2023 ACM.

关键词： Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

Natural Question Generation using Transformers and Reinforcement Learning 20

Natural Question Generation using Transformers and Reinforce...

引用

20th OITS International Conference on Information Technology, OCIT 2022

作者： Biswas, Dipshikha Nadipalli, Suneel Sneha, B. Gupta, Deepa Amudha, J. Department of Computer Science and Engineering Amrita School of Computing Bengaluru Amrita Vishwa Vidyapeetham India

ISBN: (纸本)9781665493482

Natural Question Generation (NQG) is among the most popular open research problems in Natural Language Processing (NLP) alongside Neural Machine Translation, Open Domain Chatbots, etc. Among the many approaches taken up to solve this problem, neural networks have been deemed the benchmark in this particular research area. This paper aims at adopting a generator - evaluator framework in a neural network architecture to allow additional focus on the context of the content used for framing a question. The generator uses NLP architectures like transformers (T5) to generate a question given a context while the evaluator uses Reinforcement Learning (RL) to check the correctness of the generated question. The involvement of RL has improved the results (as shown in Table 2), and there is increased computational efficiency as the training is coupled with the policy of RL. This turns the problem into a reinforcement learning task and allows for the generation of a wide range of questions for the same context-answer pair. The given algorithm is tested on the benchmark dataset - SQuAD with BLEU score as the evaluation metric © 2022 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Deep learning approach for Parkinson’s screening with geometric features from spiral and wave drawings

引用

Multimedia Tools and Applications 2025年 1-29页

作者： Malik, Meenakshi Onyema, Edeh Michael Uddin, Mueen Yadav, Poonam Sharma, Aanchal Jose, Jazlyn Shankar, Achyut Alasim, Fahad Abidi, Mustufa Haider BML Munjal University Gurugram India Department of Mathematics and Computer Science Coal City University Enugu Nigeria Adjunct Faculty Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Chennai602105 India College of Computing & Information Technology University of Doha for Science and Technology Doha Qatar Starex University Gurugram India School of Computer Science Engineering and Technology Bennett University Greater Noida Uttar Pradesh201310 India Industrial Engineering Department College of Engineering King Saud University P. O. Box-800 Riyadh11421 Saudi Arabia Advanced Manufacturing Institute King Saud University Riyadh11421 Saudi Arabia

Parkinson’s disease (PD) is the second most common neurological ailment and affects a substantial percentage of the elderly population. A deterioration in handwriting is a common indication of Parkinson’s disease. Researchers have developed an algorithm-based tool to assist in distinguishing between Parkinson’s disease patients and healthy persons because the disease can be difficult to identify. Unfortunately, previous studies have not tackled this problem or adopted a preventive approach. Examining handwriting is one way to detect PD. After analyzing multiple publications, we examined the 2020 and 2022 articles from Convolutional Neural Network utilizing deep learning techniques. Spiral and wave photos of those in good health and those with Parkinson’s illness were employed in our collection. This investigation used the Random Forest, VGG-16, and ResNet-50 approaches. We applied ResNet-50 for the drawings and got 97% accuracy for spiral pictures over 10 epochs and 98% accuracy for wave images over eight epochs by using feature extraction and preprocessing techniques. This exceptional performance demonstrates ResNet-50’s potential as a potent tool for medical diagnostics, utilizing its deep learning skills to generate precise and reliable results compared to existing studies in which ResNet-50 has 83% accuracy. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Neurodegenerative diseases

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：