检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Jarvis, Devon Lee, Sebastian Dominé, Clémentine Carla Juliette Saxe, Andrew M. Mannelli, Stefano Sarao School of Computer Science and Applied Mathematics University of the Witwatersrand South Africa Center for Computational Neuroscience Flatiron Institute Simons Foundation Gatsby Computational Neuroscience Unit Sainsbury Wellcome Centre UCL United Kingdom Data Science and AI Computer Science and Engineering Chalmers University of Technology University of Gothenburg Sweden Machine Intelligence and Neural Discovery Institute University of the Witwatersrand South Africa CIFAR Azrieli Global Scholar CIFAR

Prior work has demonstrated a consistent tendency in neural networks engaged in continual learning tasks, wherein intermediate task similarity results in the highest levels of catastrophic interference. This phenomenon is attributed to the network’s tendency to reuse learned features across tasks. However, this explanation heavily relies on the premise that neuron specialisation occurs, i.e. the emergence of localised representations. Our investigation challenges the validity of this assumption. Using theoretical frameworks for the analysis of neural networks, we show a strong dependence of specialisation on the initial condition. More precisely, we show that weight imbalance and high weight entropy can favour specialised solutions. We then apply these insights in the context of continual learning, first showing the emergence of a monotonic relation between task-similarity and forgetting in non-specialised networks. Finally, we show that specialization by weight imbalance is beneficial on the commonly employed elastic weight consolidation regularisation technique. Copyright © 2025, The Authors. All rights reserved.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

Comparison of Palm Oil Fresh Fruit Bunches (FFB) Ripeness Classification Technique using Deep Learning Method 13

Comparison of Palm Oil Fresh Fruit Bunches (FFB) Ripeness Cl...

引用

13th Asian Control Conference, ASCC 2022

作者： Khamis, Nurulaqilla Selamat, Hazlina Ghazalli, Shuwaibatulaslamiah Md Saleh, Nurul Izrin Yusoff, Nooraini Universiti Malaysia Kelantan Department of Data Science Kelantan City Campus Pengkalan Chepa Kota Bharu16100 Malaysia Universiti Teknologi Malaysia Centre for Artificial Intelligence and Robotics School of Electrical Engineering Faculty of Engineering Johor 81310 Malaysia Universiti Teknologi Malaysiam Jalan Sultan Yahya Petra Malaysia-Japan International Institute of Technology Kuala Lumpur54100 Malaysia

ISBN: (纸本)9788993215236

The ripeness of palm oil fruit is currently determined through manual visual inspection by palm oil estate workers that could result inconsistent and inaccurate fruit grading. Moreover, the manual inspection is time-consuming and exhausting duty for humans to complete the daily repetitive task. To overcome this issue, this paper proposes an automatic fruit grading classification by utilizing computer vision technologies. A comparison using image classification (ResNet50) and object detection (YOLOv3) technique is analysed in this work. It is clearly demonstrated that object detection model is remarkable in improving ripeness category based on the finer level of feature that has been extracted during the convolutional process. © 2022 ACA.

关键词： Deep learning Visualization Oils Object detection Manuals Inspection Feature extraction

来源：评论

学校读者我要写书评

暂无评论

RFiDARFusion: Enhancing Contactless Activity Monitoring with Radar and RFID Fusion

RFiDARFusion: Enhancing Contactless Activity Monitoring with...

引用

Antennas and Propagation & USNC/URSI National Radio science Meeting

作者： Muhammad Zakir Khan Muhammad Bilal Alexander Maniangat Luke Kamran Arshad Khaled Assaleh Syed Tariq Shah Muhammad Ali Imran Qammer H. Abbasi James Watt School of Engineering University of Glasgow Glasgow UK Department of Electrical Engineering National University of Technology (NUTECH) Islamabad Pakistan Department of Clinical Sciences College of Dentistry Ajman University UAE Artificial Intelligence Research Centre Aiman University Ajman UAE Department of Electrical and Computer Engineering College of Engineering and IT Ajman University UAE School of Computer Science and Electronic Engineering University of Essex Colchester UK

ISBN: (数字)9798350369908

ISBN: (纸本)9798350369915

Indoor human activity recognition (HAR) faces challenges due to the limitations of single-sensor systems in terms of privacy and cost. Our contactless RFiDARFusion system aims to reduce healthcare costs by swiftly identifying unforeseen activities through artificial intelligence and wireless communication. The system accurately discerns complex human activities by employing the LSTM-VAE fusion algorithm for multi-sensor data extraction, proving effective in long-range and non-line-of-sight environments. Two data fusion algorithms were studied across five distinct activities, significantly improving HAR accuracy by a minimum of 4.5% and 6.5% under data and feature-level fusion, respectively. The resulting RFiDARFusion system is contactless, precise, and robust, promising to enhance the quality of life for elderly patients in assisted living.

关键词： Wireless communication Privacy Costs Accuracy Radar Radar antennas Assisted living

来源：评论

学校读者我要写书评

暂无评论

Multi-granularity prior networks for uncertainty-informed patient-specific quality assurance

引用

computers in Biology and Medicine 2024年 179卷 108925页

作者： Zeng, Xiaoyang Zhu, Qizhen Ahmed, Awais Hanif, Muhammad Hou, Mengshu Jie, Qiu Xi, Rui Shah, Syed Attique School of Computer Science and Engineering University of Electronic Science and Technology of China - UESTC Sichuan611731 China Department of Radiation Oncology Peking Union Medical College Hospital Chinese Academy of Medical Sciences and Peking Union Medical College Beijing China School of Big Data and Artificial Intelligence Chengdu Technological University Sichuan611730 China School of Computing and Digital Technology Birmingham City University STEAMhouse BirminghamB4 7RQ United Kingdom

Deep Learning Automated Patient-Specific Quality Assurance (PSQA) aims to reduce clinical resource requirements. It is vital to ensure the safety and effectiveness of radiation therapy by predicting the dose difference metric (Gamma passing rate) and its distribution. However, current research overlooks uncertainty quantification in model predictions, limiting their trustworthiness in real clinical environments. This paper proposes a Multi-granularity Uncertainty Quantification (MGUQ) framework. A Bayesian framework that quantifies uncertainties at multiple granularities for multi-task PSQA, specifically Gamma Passing Rate (GPR) prediction and Dose Difference Prediction (DDP), integrates visualization-based interactive components. Using Bayesian theory, we derive a comprehensive multi-granularity loss function that comprises granularity-specific loss and coherence loss components. Additionally, we proposed Multi-granularity Prior Networks, a dual-stream network architecture, to infer the distributions of DDP (modeled as t-distributions) and GPR (modeled as Gaussian distributions) under specific statistical assumptions. Comprehensive evaluations are conducted on a dataset from ‘‘Peeking Union Medical College Hospital’’, and results show that our proposed method achieves a minimum MAE loss of 0.864 with a 2%/3 mm criterion and realizes the uncertainty visualization of dose difference. Further, it also achieves 100% Clinical Accuracy (CA) with a workload of 67.2%. Experiments demonstrate that the proposed framework can enhance the trustworthiness of deep learning applications in PSQA. © 2024

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Identifying Sources of Soil Pollution in Industrial Areas Using IoT data and Neural Networks

Identifying Sources of Soil Pollution in Industrial Areas Us...

引用

Intelligent Technologies (ACOIT), Asian Conference on

作者： Rajesh Kambattan Kovarasan Dr N V Shibu A. Sairam K. K. Manivannan M.V Ishwarya T. R. GaneshBabu Department of Computer Science and Engineering Vel Tech Rangarajan Dr. Sagunthala R & D Institute of Science and Technology Chennai Tamil Nadu India Department of Computing Technologies School of Computing SRM Institute of Science and Technology Kattankulathur Chennai Tamil Nadu India Department of Computer Science and Engineering Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Chennai Tamil Nadu India Department of Mechatronics Engineering KCG College of Technology Chennai Tamil Nadu India Artificial Intelligence and Data science Department Agni College of Technology Chennai Tamil Nadu India Department of Electronics and Communication Engineering Muthayammal Engineering College (Autonomous) Namakkal Tamil Nadu India

ISBN: (数字)9798350374957

ISBN: (纸本)9798350374964

Quick and precise source identification is crucial for developing successful remediation solutions to address soil contamination in industrial locations, which presents serious dangers to both the environment and human health. To locate the origins of soil contamination, this research presents a new method that combines neural network models with data collected from the Internet of Things (IOT). To monitor soil type, pollution levels, and weather conditions in real-time, set up IoT devices. After that, advanced neural network algorithms are used to process and analyze the data to find connections and patterns that might indicate the origins of the pollution. Our findings show that this method is useful for identifying where soil pollution inside industrial zones is coming from. Achieve high accuracy in source identification by integrating IoT technology and machine learning approaches. This allows us to address environmental harm and protect public health via focused remediation activities. Policymakers, environmental scientists, and stakeholders engaged in sustainable development and pollution control may benefit greatly which highlight the potential of IoT data and neural networks in challenges difficult environmental concerns.

关键词： Technological innovation Pollution Environmentally friendly manufacturing techniques Neural networks Pollution control Soil Internet of Things Stakeholders Soil pollution Contamination

来源：评论

学校读者我要写书评

暂无评论

Routing of data Between the Nodes in Mobile Adhoc Networks using Machine Learning Modelling

Routing of Data Between the Nodes in Mobile Adhoc Networks u...

引用

2022 International Conference on Electronics and Renewable Systems, ICEARS 2022

作者： Ramesh Babu, P Ranjan Nayak, Nihar Meenakshi, K Das, Saswata Selvakanmani, S. Ananda Kumar, K S Wollega University College of Engineering Technology Department of Computer Science Nekemte Ethiopia Department of Mca Andhra Pradesh Chittoor India Department of Mathematics Bengaluru India Cognizant Technology Solutions India Pvt. Ltd. West Bengal Kolkata India Velammal Institute of Technology Department of Artificial Intelligence and Data Science Tamil Nadu Chennai India Dilla University School of Computing and Informatics College of Engineering and Technology Snnpr Dilla Ethiopia

ISBN: (纸本)9781665484251

The routing of data between the packets in mobile adhoc networks has severe limitations that involves communication and computational overhead, which has to be reduced for effective transmission of data packets. In this paper, we develop a machine learning assisted routing mechanism that has three different modules to serve the purpose of reduced overhead during routing. The study uses genetic algorithm to resolve the routing constraints between the source and destination nodes. The modelling is specifically designed to overcome the limitations associated with the generation of overheads. The simulation is conducted to test the efficacy of the models against various existing mechanisms. The results of simulation shows that the proposed method achieves higher rate of packet delivery rate, throughput and reduced delay. © 2022 IEEE.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Hybrid Inception Architecture with Residual Connection: Fine-tuned Inception-ResNet Deep Learning Model for Lung Inflammation Diagnosis from Chest Radiographs

引用

Procedia computer science 2024年 235卷 1841-1850页

作者： Mehdi Neshat Muktar Ahmed Hossein Askari Menasha Thilakaratne Seyedali Mirjalili Center for Artificial Intelligence Research and Optimisation Torrens University Australia Brisbane QLD 4006 Australia South Australian Health and Medical Research Institute (SAHMRI) University of South Australia Adelaide SA 5000 Australia Australian Centre for Precision Health the University of South Australia Cancer Research Institute Adelaide SA 5000 Australia School of Electrical Engineering and Computer Science University of Queensland St Lucia Brisbane QLD 4072 Australia School of Computer Science The University of Adelaide Adelaide 5005 Australia

Diagnosing lung inflammation, particularly pneumonia, is of paramount importance for effectively treating and managing the disease. Pneumonia is a common respiratory infection caused by bacteria, viruses, or fungi and can indiscriminately affect people of all ages. As highlighted by the World Health Organization (WHO), this prevalent disease tragically accounts for a substantial 15% of global mortality in children under five years of age. This article presents a comparative study of the Inception-ResNet deep learning model’s performance in diagnosing pneumonia from chest radiographs. The study leverages Mendeley’s chest X-ray images dataset, which contains 5856 2D images, including both Viral and Bacterial Pneumonia X-ray images. The Inception-ResNet model is compared with seven other state-of-the-art convolutional neural networks (CNNs), and the experimental results demonstrate the Inception-ResNet model’s superiority in extracting essential features and saving computation runtime. Furthermore, we examine the impact of transfer learning with fine-tuning in improving the performance of deep convolutional models. This study provides valuable insights into using deep learning models for pneumonia diagnosis and highlights the potential of the Inception-ResNet model in this field. In classification accuracy, Inception-ResNet-V2 showed superior performance compared to other models, including ResNet152V2, MobileNet-V3 (Large and Small), EfficientNetV2 (Large and Small), InceptionV3, and NASNet-Mobile, with substantial margins. It outperformed them by 2.6%, 6.5%, 7.1%, 13%, 16.1%, 3.9%, and 1.6%, respectively, demonstrating its significant advantage in accurate classification.

关键词： Lung Infammation Pneumonia Machine learning Deep learning Inception-Resnet Fine-tuning

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Representation Learning for Time Series: A Review

arXiv

引用

arXiv 2023年

作者： Meng, Qianwen Qian, Hangwei Liu, Yong Xu, Yonghui Shen, Zhiqi Cui, Lizhen The School of Software Shandong University Jinan China Joint SDU-NTU Centre for Artificial Intelligence Research Shandong University Jinan China A*STAR Singapore Nanyang Technological University Singapore The School of Computer Science and Engineering Nanyang Technological University Singapore

Unsupervised representation learning approaches aim to learn discriminative feature representations from unlabeled data, without the requirement of annotating every sample. Enabling unsupervised representation learning is extremely crucial for time series data, due to its unique annotation bottleneck caused by its complex characteristics and lack of visual cues compared with other data modalities. In recent years, unsupervised representation learning techniques have advanced rapidly in various domains. However, there is a lack of systematic analysis of unsupervised representation learning approaches for time series. To fill the gap, we conduct a comprehensive literature review of existing rapidly evolving unsupervised representation learning approaches for time series. Moreover, we also develop a unified and standardized library, named ULTS (i.e., Unsupervised Learning for Time Series), to facilitate fast implementations and unified evaluations on various models. With ULTS, we empirically evaluate state-of-the-art approaches, especially the rapidly evolving contrastive learning methods, on 9 diverse real-world datasets. We further discuss practical considerations as well as open research challenges on unsupervised representation learning for time series to facilitate future research in this field. © 2023, CC BY.

关键词： Time series

来源：评论

学校读者我要写书评

暂无评论

AI based Mobile Machine Learning Framework for Edge Video Analytics

AI based Mobile Machine Learning Framework for Edge Video An...

引用

Intelligent & Innovative Practices in engineering & Management (IIPEM), International Conference on

作者： Kamalesh Thoppaen Suresh Babu Tarun Dhar Diwan G. Prasanna Lakshmi Lakshmi Narasimhan Bhagyashree Ashok Tingare Pravin B Khatkale Globallogic Inc Santa Clara CA USA Atal Bihari Vajpayee University Bilaspur India School of Computer Science and Engineering Sandip University Nashik Maharashtra Krishnamoorthy Senior Technical Lead HCL America Solutions Inc Sunnyvale California Department of Artificial Intelligence and Data Science D Y Patil College of Engineering Akurdi Pune Sanjivani University Kopargaon Maharashtra India

ISBN: (数字)9798350390049

ISBN: (纸本)9798350390056

The rapid advancements in mobile technology and edge computing have spurred the development of innovative video analytics solutions that leverage the computational power of mobile devices for real-time processing and decision-making. This paper presents an AI-based mobile machine learning framework tailored for edge video analytics, addressing the challenges of latency, bandwidth consumption, and privacy inherent in traditional cloud-based approaches. The proposed framework integrates advanced machine learning models optimized for mobile hardware, utilizing techniques such as model quantization, pruning, and hardware acceleration to deliver efficient and accurate video analytics directly on edge devices. Experimental results demonstrate the framework's effectiveness across various mobile platforms, showing significant improvements in latency reduction, battery efficiency, and data privacy compared to cloud-based solutions. The framework's scalability and adaptability make it a powerful tool for deploying sophisticated video analytics in diverse edge computing environments.

关键词： Analytical models Privacy Visual analytics Computational modeling Image edge detection Surveillance Machine learning Real-time systems Mobile handsets Edge computing

来源：评论

学校读者我要写书评

暂无评论

Physics-informed Deep Learning to Solve Three-dimensional Terzaghi's Consolidation Equation: Forward and Inverse Problems

arXiv

引用

arXiv 2024年

作者： Yuan, Biao Heitor, Ana Wang, He Chen, Xiaohui Geomodelling and Artificial Intelligence Centre School of Civil Engineering University of Leeds LeedsLS2 9JT United Kingdom Department of Computer Science Faculty of Engineering Sciences University College London LondonWC1E 6BT United Kingdom

The emergence of neural networks constrained by physical governing equations has sparked a new trend in deep learning research, which is known as Physics-Informed Neural Networks (PINNs). However, solving high-dimensional problems with PINNs is still a substantial challenge, the space complexity brings difficulty to solving large multidirectional problems. In this paper, a novel PINN framework to quickly predict several three-dimensional Terzaghi's consolidation cases under different conditions is proposed. Meanwhile, the loss functions for different cases are introduced, and their differences in three-dimensional consolidation problems are highlighted. The tuning strategies for the PINNs framework for three-dimensional consolidation problems are introduced. Then, the performance of PINNs is tested and compared with traditional numerical methods adopted in forward problems, and the coefficients of consolidation and the impact of noisy data in inverse problems are identified. Finally, the results are summarized and presented from three-dimensional simulations of PINNs, which show an accuracy rate of over 99% compared with ground truth for both forward and inverse problems. These results are desirable with good accuracy and can be used for soil settlement prediction, which demonstrates that the proposed PINNs framework can learn the three-dimensional consolidation PDE well. Copyright © 2024, The Authors. All rights reserved.

关键词： Inverse problems

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：