检索结果-内蒙古大学图书馆

Proceedings of the 41st International Conference on Machine Learning

作者： Chenxiao Yang Qitian Wu David Wipf Ruoyu Sun Junchi Yan School of Artificial Intelligence & Department of Computer Science and Engineering & MoE Lab of AI Shanghai Jiao Tong University Amazon Web Services School of Data Science The Chinese University of Hong Kong Shenzhen and Shenzhen International Center for Industrial and Applied Mathematics Shenzhen Research Institute of Big Data

A long-standing goal in deep learning has been to characterize the learning behavior of black-box models in a more interpretable manner. For graph neural networks (GNNs), considerable advances have been made in formalizing what functions they can represent, but whether GNNs will learn desired functions during the optimization process remains less clear. To fill this gap, we study their training dynamics in function space. In particular, we find that the gradient descent optimization of GNNs implicitly leverages the graph structure to update the learned function, as can be quantified by a phenomenon which we call kernel-graph alignment. We provide theoretical explanations for the emergence of this phenomenon in the overparameterized regime and empirically validate it on real-world GNNs. This finding offers new interpretable insights into when and why the learned GNN functions generalize, highlighting their limitations in heterophilic graphs. Practically, we propose a parameter-free algorithm that directly uses a sparse matrix (i.e. graph adjacency) to update the learned function. We demonstrate that this embarrassingly simple approach can be as effective as GNNs while being orders-of-magnitude faster.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Effect of Sliding Window Sizes on Sensor-Based Human Activity Recognition Using Smartwatch Sensors and Deep Learning Approaches

Effect of Sliding Window Sizes on Sensor-Based Human Activit...

引用

Big Data Analytics and Practices (IBDAP), International Conference on

作者： Sakorn Mekruksavanich Ponnipa Jantawong Anuchit Jitpattanakul Department of Computer Engineering School of Information and Communication Technology University of Phayao Phayao Thailand Department of Mathematics Faculty of Applied Science Intelligent and Nonlinear Dynamic Innovations Research Center King Mongkut's University of Technology North Bangkok Bangkok Thailand

ISBN: (数字)9798350391749

ISBN: (纸本)9798350391756

Smartwatch sensors for human activity recognition (HAR) have gained significant attention due to their applications in healthcare and fitness monitoring. The effectiveness of HAR systems largely depends on the choice of sliding window widths for sensor data segmentation. This study investigates the impact of varying sliding window widths on the accuracy of HAR using wristwatch sensors and deep learning techniques. We conducted experiments using the daily human activity (DHA) dataset, comprising sensor data from 11 distinct activities. Data was preprocessed and segmented using window sizes ranging from 5 to 40 seconds. Four deep learning models (CNN, LSTM, BiLSTM, and CNN-LSTM) were employed and evaluated using accuracy, precision, recall, and F1-score. Window size significantly affected HAR performance. Smaller windows improved short-duration activity recognition but increased computational complexity, while larger windows reduced computational load but decreased accuracy for rapid activity changes. The CNN-LSTM hybrid model consistently outperformed other models, achieving 92.11% accuracy with a 20-second window and overlapping segmentation. This research provides valuable insights into balancing recognition accuracy and computational resources in smartwatch sensor-based HAR, contributing to the development of efficient and accurate systems for real-world applications.

关键词： Deep learning Wearable Health Monitoring Systems Accuracy Computational modeling Medical services Sensors Human activity recognition

来源：评论

学校读者我要写书评

暂无评论

Cybersecurity management in cloud computing: semantic literature review and conceptual framework proposal

引用

Journal of Reliable Intelligent Environments 2021年第2期7卷 69-84页

作者： Tissir, Najat El Kafhali, Said Aboutabit, Noureddine Process Engineering Computer Science and Mathematics Laboratory National School of Applied Sciences Sultan Moulay Slimane University Khouribga25000 Morocco Sultan Moulay Slimane University Beni Mellal Morocco Hassan First University of Settat Faculty of Sciences and Techniques Computer Networks Mobility and Modeling Laboratory: IR2M Settat26000 Morocco

Cloud Computing is an emerging paradigm that is based on the concept of distributed computing. Its definition is related to the use of computer resources which are offered as a service. As with any novel technology, Cloud Computing is subject to security threats, vulnerabilities, and attacks. Recently, the studies on security impact include the interaction of software, people and services on the Internet and that is called cyber-security or cyberspace security. In spite of various studies, we still fail to define the needs of cybersecurity management in Cloud Computing. This paper principally focuses on a comprehensive study of Cloud Computing concerns, security, cybersecurity differences, ISO, and NIST standards. It aims at identifying the policies and the guidelines included in these standards as well as it provides a comprehensive Framework proposal to manage and prevent cyber risks in Cloud Computing taking into consideration the ISO 27,032, ISO 27,001, ISO 27,017 and NIST cybersecurity Framework CSF. In addition to that, our study pinpoints at the criteria that concern measuring the maturity of organizations that implement the framework. Our objective is to provide guidance to organizations on how to establish their proper approach of cybersecurity risk management in Cloud Computing or to complement their ‘already have’ processes. © 2020, Springer Nature Switzerland AG.

关键词： Cybersecurity

来源：评论

学校读者我要写书评

暂无评论

Complete Neural Networks for Complete Euclidean Graphs

arXiv

引用

arXiv 2023年

作者： Hordan, Snir Amir, Tal Gortler, Steven J. Dym, Nadav Faculty of Mathematics Technion – Israel Institute of Technology Haifa Israel School of Engineering and Applied Sciences Harvard University Cambridge United States Faculty of Computer Science Technion – Israel Institute of Technology Haifa Israel

Neural networks for point clouds, which respect their natural invariance to permutation and rigid motion, have enjoyed recent success in modeling geometric phenomena, from molecular dynamics to recommender systems. Yet, to date, no model with polynomial complexity is known to be complete, that is, able to distinguish between any pair of non-isomorphic point clouds. We fill this theoretical gap by showing that point clouds can be completely determined, up to permutation and rigid motion, by applying the 3-WL graph isomorphism test to the point cloud’s centralized Gram matrix. Moreover, we formulate an Euclidean variant of the 2-WL test and show that it is also sufficient to achieve completeness. We then show how our complete Euclidean WL tests can be simulated by an Euclidean graph neural network of moderate size and demonstrate their separation capability on highly symmetrical point clouds. © 2023, CC BY.

关键词： Molecular dynamics

来源：评论

学校读者我要写书评

暂无评论

Comparison of Different Deep Learning Networks to Classify Epilepsy Seizure Based on EEG Signals

Comparison of Different Deep Learning Networks to Classify E...

引用

International Conference on Electrical engineering/Electronics, computer, Telecommunications and Information Technology, ECTI-CON

作者： Sakorn Mekruksavanich Ponnipa Jantawong Anuchit Jitpattanakul Department of Computer Engineering School of Information and Communication Technology University of Phayao Phayao Thailand Intelligent and Nonlinear Dynamic Innovations Research Center Department of Mathematics Faculty of Applied Science King Mongkut's University of Technology North Bangkok Bangkok Thailand

ISBN: (数字)9798350381559

ISBN: (纸本)9798350381566

Epilepsy, a widespread neurological condition impacting approximately 50 million people worldwide, is often identified and studied through electroencephalography (EEG) due to its rich temporal information. Convolutional neural networks and recurrent neural networks have demonstrated potential in automatically detecting seizures from EEG data. Despite this, there's a shortage of comprehensive studies comparing the detection capabilities of different designs. This research assesses and contrasts the effectiveness of five renowned deep neural network models - CNN, LSTM, BiLSTM, GRU, and BiGRU - in binary categorization of EEG segments into seizure and non-seizure groups. Training and evaluation utilize the publicly accessible Bonn dataset, specifically crafted for epileptic seizure detection. Performance metrics like accuracy, precision, recall, and F1-score are employed in the analysis. The principal findings highlight BiGRU's impressive accuracy at 95.80% and F1-score at 92.96% in individualized seizure identification. Additionally, the study introduces a groundbreaking deep learning architecture, the self-normalizing neural network (SNN), to enhance seizure detection accuracy. The recommended SNN exhibits outstanding performance, boasting an accuracy of 98.50% and an impressive F1-score of 98.50%.

关键词： Training Measurement Accuracy Recurrent neural networks Epilepsy Electroencephalography Telecommunications

来源：评论

学校读者我要写书评

暂无评论

A Comprehensive Evaluation of Driver Drowsiness Identification System using Camera based Improved Deep Learning Methodology 9

A Comprehensive Evaluation of Driver Drowsiness Identificati...

引用

9th IEEE International Conference on Smart Structures and Systems, ICSSS 2023

作者： Malathi, P. Sivakami Sundari, M. Karkuzhali, S. Umamaheswari, B. Thenmozhi, A. Aiyyasamy, Velu Simats Saveetha College of Liberal Arts and Sciences Department of Data Analytics Tamil Nadu Chennai India OMR Chennai Agni College of Technology Department of Mathematics Tamil Nadu Chennai India R.M.K. Engineering College Department of General Engineering Tamil Nadu Chennai India Prince Shri Venkateshwara Padmavathy Engineering College Department of Computer Science and Engineering Tamil Nadu Chennai India Kings Engineering College Department of Electronics and Communication Engineering Tamil Nadu Chennai India Meenakshi Sundararajan Engineering College Department of M.E.- Applied Electronics Tamil Nadu Chennai India

ISBN: (纸本)9798350384208

Particularly on highways, drowsy driving causes a large number of collisions while driving. With the goal to identify driver sleepiness while enhancing roadway security, it is now vital to grasp the situation and take early remedial steps. Using a proposed Image based Learning Strategy for Drowsiness Identification (ILSDI) and cross-validation with the conventional model called Convolutional Neural Network (CNN), the suggested framework offers an approach to assess the degree of fatigue among drivers according to modifications in a driver eyeballs motion. This will assist in deal with the problem concerning roadway protection. In addition, four types of expressions on the face were identified and categorized - open, closed, blinking, and no gazing using ILSDI and CNN models, indicating levels of tiredness. Finding, following, and analyzing the driver's face and eyes in real-time to calculate a sleepiness index is the goal of this technology, which operates in different lighting circumstances. Avoiding these kinds of accidents is possible with the help of a driver sleepiness monitoring structure, which uses a digital camera and accompanying software to measure the rate of blinking as well as the dimension of the driver's eyes. The driver sleepiness recognition system can identify when the driver is getting sleepy and sound an alarm if necessary. It is based on an offline implementation of a deep learning algorithm that uses ILSDI and CNN. © 2023 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Phy-Diff: Physics-guided Hourglass Diffusion Model for Diffusion MRI Synthesis

arXiv

引用

arXiv 2024年

作者： Zhang, Juanhua Yan, Ruodan Perelli, Alessandro Chen, Xi Li, Chao School of Science and Engineering University of Dundee United Kingdom Department of Applied Mathematics and Theoretical Physics University of Cambridge United Kingdom Department of Computer Science University of Bath United Kingdom Department of Clinical Neurosciences University of Cambridge United Kingdom School of Medicine University of Dundee United Kingdom

Diffusion MRI (dMRI) is an important neuroimaging technique with high acquisition costs. Deep learning approaches have been used to enhance dMRI and predict diffusion biomarkers through undersampled dMRI. To generate more comprehensive raw dMRI, generative adversarial network based methods are proposed to include b-values and b-vectors as conditions, but they are limited by unstable training and less desirable diversity. The emerging diffusion model (DM) promises to improve generative performance. However, it remains challenging to include essential information in conditioning DM for more relevant generation, i.e., the physical principles of dMRI and white matter tract structures. In this study, we propose a physics-guided diffusion model to generate high-quality dMRI. Our model introduces the physical principles of dMRI in the noise evolution in the diffusion process and introduces a query-based conditional mapping within the diffusion model. In addition, to enhance the anatomical fine details of the generation, we introduce the XTRACT atlas as a prior of white matter tracts by adopting an adapter technique. Our experiment results show that our method outperforms other state-of-the-art methods and has the potential to advance dMRI enhancement. © 2024, CC BY-NC-ND.

关键词： Magnetic resonance imaging

来源：评论

学校读者我要写书评

暂无评论

Boosting the oxygen reduction activity on metal surfaces by fine-tuning interfacial water with midinfrared stimulation

引用

The Innovation 2025年第1期6卷 80-86,79页

作者： Qilin Zhang Yu Wu Hao Sun Zhongjie Zhu Hongwei Zhao Jinrong Yang Jie Wang Min Chen Sanzhao Song Shiyou Zheng Dengsong Zhang Hui Yang Zhi Zhu Chunlei Wang School of Mathematics Physics and Finance Anhui Polytechnic UniversityWuhu 241000China Shanghai Institute of Applied Physics Chinese Academy of SciencesShanghai 201800China School of Materials and Chemistry University of Shanghai for Science and TechnologyShanghai 200093China Shanghai Advanced Research Institute Chinese Academy of SciencesShanghai 201210China School of Chemistry and Molecular Engineering East China Normal UniversityShanghai 200062China Wenzhou Institute University of Chinese Academy of SciencesWenzhouZhejiang 325001China Innovation Institute of Carbon Neutrality Shanghai UniversityShanghai 200444China College of Medical Instrumentation Shanghai University of Medicine and Health SciencesShanghai 201318China School of Optical-Electrical and Computer Engineering University of Shanghai for Science and TechnologyShanghai 200093China International Joint Laboratory of Catalytic Chemistry College of SciencesShanghai UniversityShanghai 200444China

Heterogeneous catalysis at the metal surface generally involves the transport of molecules through the interfacial water layer to access the surface,which is a rate-determining step at the *** this study,taking the oxygen reduction reaction on a metal electrode in aqueous solution as an example,using accurate molecular dynamic simulations,we propose a novel long-range regulation strategy in which midinfrared stimulation(MIRS)with a frequency of approximately 1,000 cm^(-1)is applied to nonthermally induce the structural transition of interfacial water from an ordered to disordered state,facilitating the access of oxygen molecules to metal surfaces at room temperature and increasing the oxygen reduction activity ***,the theoretical prediction is confirmed by the experimental observation of a significant discharge voltage increase in zinc-air batteries under *** MIRS approach can be seamlessly integrated into existing strategies,offering a new approach for accelerating heterogeneous reactions and gas sensing within the interfacial water system.

关键词： accurate molecular dynamic simulationswe heterogeneous catalysis metal electrode interfacial water metal surface transport molecules interfacial water layer midinfrared stimulation midinfrared stimulation mirs

来源：评论

学校读者我要写书评

暂无评论

Rd-Dip: Rician Denoising Deep Image Prior

SSRN

引用

SSRN 2025年

作者： Iglesias-Goldaracena, Héctor Rodrigo Díaz, Iván Ramírez Schiavi, Emanuele Applied Mathematics Materials Science and Engineering and Electronic Technology Department Universidad Rey Juan Carlos Calle Tulipán s/n Móstoles28933 Spain Computer Science and Statistics Department Universidad Rey Juan Carlos Calle Tulipán s/n Móstoles28933 Spain

Rician denoising of magnetic resonance images (MRI) is a fundamental problem in medical image processing. Although variational methods can address Rician noise, the mathematical complexity of the underlying models makes them challenging to implement. Neural networks offer a compelling alternative but require extensive labelled datasets and are prone to introducing artifacts. This study proposes advanced tailored models for MRI Rician denoising within the unsupervised Deep Image Prior (DIP) framework, eliminating the need for additional data and reducing training-related artifacts. To our knowledge, this is the first application of the Rician maximum likelihood in the DIP framework. Our experiments, conducted on the Brainweb dataset, demonstrate the superiority of the proposed models. Additionally, we provide a detailed pipeline to ensure the reproducibility of our experiments. The code is available at https://***/heqro/rd-dip. © 2025, The Authors. All rights reserved.

关键词： Magnetic resonance imaging

来源：评论

学校读者我要写书评

暂无评论

Convolutional Neural Network with CBAM Module for Fitness Activity Recognition Using Wearable IMU Sensors

Convolutional Neural Network with CBAM Module for Fitness Ac...

引用

International ECTI Northern Section Conference on Electrical, Electronics, computer and Telecommunications engineering (ECTI-NCON)

作者： Sakorn Mekruksavanich Ponnipa Jantawong Wikanda Phaphan Anuchit Jitpattanakul Department of Computer Engineering School of Information and Communication Technology University of Phayao Phayao Thailand Department of Applied Statistics Faculty of Applied Science Research Group in Statistical Learning and Inference King Mongkut’s University of Technology North Bangkok Bangkok Thailand Department of Mathematics Faculty of Applied Science Intelligent and Nonlinear Dynamic Innovations Research Center King Mongkut’s University of Technology North Bangkok Bangkok Thailand

Wearable motion sensors are gaining traction for fitness activity tracking due to their utility in assisted living, wellness tracking, and fitness training applications. Such sensors typically incorporate accelerometers and gyroscopes to monitor and log physical activities. This work proposes a deep-learning approach for identifying fitness activities using wearable inertial measurement units (IMUs) signals. Specifically, a convolutional neural network (CNN) integrated with a convolutional block attention module (CBAM) provides the core framework. The CBAM module applies channel and spatial attention mechanisms to enhance CNN’s capacity to discriminate salient features. Multichannel IMU time-series data, comprising accelerometer and gyroscope readings as subjects perform various exercises, supply input to the model. By learning via an attention-enhanced architecture, the model can extract robust and informative representations to categorize activities accurately. Experiments on a public IMU dataset with five exercise types across over twenty participants evaluated the approach. Results showed over 99.58% subject-independent activity recognition accuracy when augmenting the CNN with CBAM, constituting significant performance gains. The research findings indicate that an efficient deep-learning solution with attention modeling can recognize fitness activities reliably from on-body sensors. The proposed methodology holds substantial promise for applications in sports analytics and personalized health tracking.

关键词：

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：