检索结果-内蒙古大学图书馆

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Yusong Wang Chaoran Cheng Shaoning Li Yuxuan Ren Bin Shao Ge Liu Pheng-Ann Heng Nanning Zheng National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications and Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University University of Illinois Urbana-Champaign Department of Computer Science and Engineering The Chinese University of Hong Kong University of Science and Technology of China Microsoft Research AI4Science

ISBN: (纸本)9798331314385

Geometric graph neural networks (GNNs) have emerged as powerful tools for modeling molecular geometry. However, they encounter limitations in effectively capturing long-range interactions in large molecular systems due to the localization assumption of GNN. To address this challenge, we introduce Neural P3M, a versatile enhancer of geometric GNNs to expand the scope of their capabilities by incorporating mesh points alongside atoms and reimaging traditional mathematical operations in a trainable manner. Neural P3M exhibits flexibility across a wide range of molecular systems and demonstrates remarkable accuracy in predicting energies and forces, outperforming on benchmarks such as the MD22 dataset. It also achieves an average improvement of 22% on the OE62 dataset while integrating with various architectures. Codes are available at https://***/OnlyLoveKFC/Neural_P3M.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Deep Image Clustering by Spiking Neural Networks

Deep Image Clustering by Spiking Neural Networks

引用

Applied Imagery Pattern Recognition Workshop (AIPR)

作者： Arash Mahyari Hadi AliAkbarpour Florida Institute For Human and Machine Cognition (IHMC) Pensacola FL USA Department of Intelligent Systems and Robotics University of West Florida Pensacola FL USA Department of Computer Science Saint Louis University St. Louis MO USA

Neuromorphic cameras, or event cameras, are biologically-inspired sensors that detect changes in illumination at a pixel level, different from traditional cameras where each pixel independently and asynchronously outputs when it senses illumination changes. The neuromorphic cameras exhibit high-temporal resolution and dynamic range, useful in various applications. Spiking Neural Networks (SNNs), mimicking biological neurons, are efficient in processing neuromorphic images due to their event-driven, low-power nature. However, training SNNs is challenging because of the non-differentiability of neuron models which require different optimization techniques. Inspired by recent works on unsupervised deep clustering in conventional deep learning, we introduce a novel deep clustering algorithm that employs SNNs to extract and group visual features from neuromorphic images into clusters. It utilizes an SNN feature extractor for neuromorphic images, grouping the extracted features using K-means clustering. This algorithm, which doesn’t require pre-trained SNN weights, alternates between cluster updates and SNN weight adjustments. The extracted feature vectors in SNN are clustered and used as pseudo-labels for supervised training of the SNN extractor. The performance of the proposed algorithm is evaluated on two datasets of MNIST and NMNIST, and its effectiveness is assessed using the Normalized Mutual Information (NMI) metric. The performance of our SNN system with neuromorphic images, while marginally below its counterparts of ANNs with traditional images, demonstrates considerable potential.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Enhancing Model Explainability in Financial Trading Using Training Aid Samples: A CNN-Based Candlestick Pattern Recognition Approach

Enhancing Model Explainability in Financial Trading Using Tr...

引用

IEEE International Conference on Information Reuse and Integration (IRI)

作者： Yun-Cheng Tsai Jun-Hao Chen Department of Technology Application and Human Resource Development National Taiwan Normal University Taipei Taiwan R.O.C. Department of Computer Science and Information Engineering National Taiwan University Taipei Taiwan R.O.C.

Candlestick pattern recognition is a widely adopted technique in financial trading, leveraging visual patterns to analyze price movements. Deep Convolutional Neural Networks (CNNs) have exhibited remarkable accuracy in this domain. However, the increasing demand for transparency and explainability in CNN-based models raises concerns regarding their applicability in trading decision-making. This paper addresses these concerns by presenting a framework that enhances the explainability of CNN-based candlestick pattern recognition models. Our approach introduces an innovative data augmentation method to generate training aid samples, facilitating the model’s learning process within human domains. By incorporating this framework, traders gain valuable insights into the decision-making process, comprehending the rationale behind the model’s predictions. Our proposed approach exposes the inherent “black box” nature of CNN-based models, improving their interpretability and empowering traders to make well-informed decisions based on transparent and understandable insights. This advancement holds significant potential for enhancing decision-making processes in financial trading and fostering trust among traders.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Database Management Difficulties in the Internet of Things 5

Database Management Difficulties in the Internet of Things

引用

5th International Conference on Contemporary Computing and Informatics, IC3I 2022

作者： Lourens, Melanie Tamizhselvi, A. Goswami, Brijesh Alanya-Beltran, Joel Aarif, Mohd. Gangodkar, Durgaprasad Durban University of Technology Department of Human Resources Management Durban Berea South Africa Information Technology GLA University Institute of Business Management Uttar Pradesh Mathura India Universidad Tecnológica Del Perú Department of Electronic Peru Shri JJT University Rajasthan Jhunjhunu India Graphic Era Deemed to Be University Department of Computer Science & Engineering Uttarakhand Dehradun India

ISBN: (纸本)9798350398267

The 'Internet of Things (IoT)' is an internet protocol for which real-world, virtual as well as digital objects are given recognition, detecting, connectivity, and process technology so they can interact with one another and other Internet-connected devices and services to carry out users' tasks. There are many IoT solutions available to improve and comfort civilian lives. Additionally, the use of IoT technology in the automotive sector gave rise to the concept of the 'Industrial Internet of Things (IIoT),' which has simplified the usage of Cyber Physic Systems, which enable machine and human communication. In general, the variety, heterogeneity, and vast volume of data produced by these businesses make the use of traditional database management systems inappropriate. While constructing IoT data management systems, a number of special issues should be taken into account. These varied guiding notions have led to the proposal of range Of iot data management strategies. The Internet of Things will undoubtedly become a realization as more gadgets are linked to the Internet. Massive amounts of information will be instantiated by items in environment. A high rate and numerous increments will be made to its quantity. This research paper has highlighted database management and its challenges in IoT technology through secondary qualitative analysis. © 2022 IEEE.

关键词： Big data

来源：评论

学校读者我要写书评

暂无评论

The Critical Role Played by Big Data Management in Effectively Addressing the Security and Overall Privacy Concerns Through Correlation Analysis 2

The Critical Role Played by Big Data Management in Effective...

引用

2nd International Conference on Advance Computing and Innovative Technologies in Engineering, ICACITE 2022

作者： Ahamad, Shahanawaj Roshan, Amit Lourens, Melanie Shekher, Vineet Joshi, Kapil Alanya-Beltran, Joel College of Computer Science and Engineering University of Hail Saudi Arabia Information Technology Dr Ram Manohar Lohiya Avadh University UP Ayodhya India Durban University of Technology Department of Human Resources Management South Africa Bit Sindri Electrical Engineering Department Dhanbad India UIt Uttaranchal University India Universidad Tecnologica DelPeru Professor. Electronic Department Peru

ISBN: (纸本)9781665437899

Due to the widespread application of big data in a wide range of fields, there has been a significant rise in a diverse range of data assets, and numerous data analysis technologies, such as standardized data mining or quantitative analytical techniques, are speeding up the constant enlargement of the big data industry. The fact that information from several different resources has a life cycle from acquisition to disposal is a key property of big data, as is the fact that new information may be obtained via analysis, combining, and exploitation of existing data. Yet, information security & dependability challenges arise at every stage of the life cycle, rendering the protection of individually identifying data a vital goal throughout the whole process. It is possible to evaluate user behavior using different big data analytics techniques, and the collection of this information results in the violation of personal privacy. This study analyses dangers and security challenges that arise during the administration of large amounts of data by verifying the existing standards produced by international standardization bodies and doing an analysis of relevant research. © 2022 IEEE.

关键词： Big data

来源：评论

学校读者我要写书评

暂无评论

Mixed reality based respiratory liver tumor puncture navigation

引用

Computational Visual Media 2019年第4期5卷 363-374页

作者： Ruotong Li Weixin Si Xiangyun Liao Qiong Wang Reinhard Klein Pheng-Ann Heng Institute of Computer Science II University of Bonn53115 BonnGermany Shenzhen Key Laboratory of Virtual Reality and Human Interaction Technology Shenzhen Institutes of Advanced TechnologyChinese Academy of SciencesChina Department of Computer Science and Engineering Chinese University of Hong KongHong KongChina

This paper presents a novel mixed reality based navigation system for accurate respiratory liver tumor punctures in radiofrequency ablation(RFA).Our system contains an optical see-through head-mounted display device(OST-HMD),Microsoft Holo Lens for perfectly overlaying the virtual information on the patient,and a optical tracking system NDI Polaris for calibrating the surgical utilities in the surgical *** with traditional navigation method with CT,our system aligns the virtual guidance information and real patient and real-timely updates the view of virtual guidance via a position tracking *** addition,to alleviate the difficulty during needle placement induced by respiratory motion,we reconstruct the patientspecific respiratory liver motion through statistical motion model to assist doctors precisely puncture liver *** proposed system has been experimentally validated on vivo pigs with an accurate real-time registration approximately 5-mm mean FRE and TRE,which has the potential to be applied in clinical RFA guidance.

关键词： mixed reality human computer interaction statistical motion model

来源：评论

学校读者我要写书评

暂无评论

Efficiently Fusing Sparse Lidar for Enhanced Self-Supervised Monocular Depth Estimation

Efficiently Fusing Sparse Lidar for Enhanced Self-Supervised...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Yue Wang Mingrong Gong Lei Xia Qieshi Zhang Jun Cheng Guangdong-Hong Kong-Macao Joint Laboratory of Human-Machine Intelligence-Synergy Systems Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen China Department of Computer Science University College London UK

Monocular self-supervised depth estimation with a low-cost sensor is the mainstream solution to gathering dense depth maps for robots and autonomous driving. In this paper, based on the philosophy "less is more" (i.e., focusing only on valid pixels in sparse LiDAR), we propose a novel framework, Efficient Sparse Depth (EffisDepth), for predicting dense depth. The Sparse Feature Extractor (SFE) embedded in the proposed framework effectively handles sparse LiDAR by forming sparse tensors. The Slender Group Block (SGB) is the main building block in SFE, which extracts features from sparse tensors via a structure of two branches. Extensive experiments show that our method achieves state-of-the-art performance on the KITTI benchmark, demonstrating the effectiveness of each proposed component and the self-supervised learning framework.

关键词： Laser radar Tensors Philosophical considerations Fuses Estimation Self-supervised learning Signal processing

来源：评论

学校读者我要写书评

暂无评论

FAIR Data Point:A FAIR-Oriented Approach for Metadata Publication

引用

Data Intelligence 2023年第1期5卷 163-183页

作者： Luiz Olavo Bonino da Silva Santos Kees Burger Rajaram Kaliyaperumal Mark D.Wilkinson Services and Cybersecurity group Faculty of Electrical EngineeringMathematics and Computer ScienceUniversity of Twente-Enschede7513GBthe Netherlands Biosemantics group Department of Human GeneticsLeiden University Medical Center-LeidenNL 2333 ZCthe Netherlands Escuela Técnica Superior de Ingenieria Agronomica Alimentaria y de BiosistemasCentro de Biotecnologia y Genomica de Plantas UPM-INIAUniversidad Politecnica de Madrid(UPM)-Instituto Nacional de Investigacion y Tecnologia Agraria y Alimentaria(INIA)MadridES 28223Spain

Metadata,data about other digital objects,play an important role in FAIR with a direct relation to all FAIR *** this paper we present and discuss the FAIR Data Point(FDP),a software architecture aiming to define a common approach to publish semantically-rich and machine-actionable metadata according to the FAIR *** present the core components and features of the FDP,its approach to metadata provision,the criteria to evaluate whether an application adheres to the FDP specifications and the service to register,index and allow users to search for metadata content of available FDPs.

关键词： FAIR FAIR data point FAIR principles Metadata Interoperability Linked data Semantic interoperability

来源：评论

学校读者我要写书评

暂无评论

Predicting Adolescent Female Stress with Wearable Device Data Using Machine and Deep Learning

Predicting Adolescent Female Stress with Wearable Device Dat...

引用

International Workshop on Wearable and Implantable Body Sensor Networks (BSN)

作者： Claire Jin Ame Osotsi Zita Oravecz School of Computer Science Carnegie Mellon University Pittsburgh USA Natural Language Processing Division Optum University Park USA Department of Human Development and Family Studies The Pennsylvania State University University Park USA

The prevalence of mental health issues in adolescent females has become a significant concern in recent years. To investigate the potential of wearable biosensors in predicting stress responses in this understudied demographic, we collected wearables data from eight teenage girls over 1-4 months and explored stress prediction using several machine learning (ML) and deep learning (DL) models. Various person-dependent and person-independent prediction schemes, feature extraction methods, and classifier types were systematically investigated to provide recommendations for effective stress prediction. Feature importance for the physiological signals was also analyzed to provide insights into adolescent stress responses. The study provides actionable recommendations for classifiers, feature extraction, and personalization schemes to enhance stress prediction accuracy, enhancing the understanding and early detection of mental health issues in adolescent females.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Predicting Heart Disease Using Machine Learning Techniques on Electronic Health Records Data 3

Predicting Heart Disease Using Machine Learning Techniques o...

引用

3rd International Conference on Emerging Smart Technologies and Applications, eSmarTA 2023

作者： Baddah, Waheeb Qasem, Hamzah Ali Alsabry, Ayman Hussein Obaid, Abdulrahman Mohammed Abdo Mohammed, Awadh Hanash, F.E. Azal University for Human Development Department of Information Technology Sana'a Yemen International University of Technology Twintech 21 September University of Medical and Applied Sciences Sana'a Yemen International University of Technology Twintech Department of Computer Science Sana'a Yemen Medical Information Technology 21 September University of Medical and Applied Sciences Sana'a Yemen Biomedical Equipments Technology 21 September University of Medical and Applied Sciences Sana'a Yemen International University of Technology Twintech Sana'a Yemen

ISBN: (纸本)9798350305333

heart disease is a prevalent cause of mortality worldwide, and the ability to identify and prevent this ailment at an early stage is crucial for enhancing patient outcomes. Recently, researchers have focused on utilizing Machine-Learning (ML) techniques to predict risk factors and the occurrence of heart disease. This study aims to identify the optimal parameters for accurately predicting the occurrence of heart disease through the fine-tuning of ML models' parameters. This study incorporated a blend of datasets from Cleveland, Hungary, Switzerland, and Long Beach V, which consisted of patients' medical histories and laboratory test outcomes. The collected data were employed to predict the occurrence of heart disease through diverse ML algorithms, such as KNN, decision tree, and SVM. Bayesian optimization was also utilized to fine-tune the hyperparameters of the ML algorithms for heart disease prediction. The findings demonstrated that the ML models achieved high accuracy in predicting the occurrence of heart disease, with the optimized KNN model achieving the highest accuracy rate of 100%. © 2023 IEEE.

关键词： Forecasting

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：