检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Li, Zeju Zheng, Ying-Qiu Chen, Chen Jbabdi, Saad FMRIB Centre Wellcome Centre for Integrative Neuroimaging University of Oxford Oxford United Kingdom Department of Computer Science University of Sheffield Sheffield United Kingdom Department of Engineering Science University of Oxford Oxford United Kingdom Department of Computing Imperial College London London United Kingdom

Semi-supervised learning (SSL) algorithms struggle to perform well when exposed to imbalanced training data. In this scenario, the generated pseudo-labels can exhibit a bias towards the majority class, and models that employ these pseudo-labels can further amplify this bias. Here we investigate pseudo-labeling strategies for imbalanced SSL including pseudo-label refinement and threshold adjustment, through the lens of statistical analysis. We find that existing SSL algorithms which generate pseudo-labels using heuristic strategies or uncalibrated model confidence are unreliable when imbalanced class distributions bias pseudo-labels. To address this, we introduce SEmi-supervised learning with pseudo-label optimization based on VALidation data (SEVAL) to enhance the quality of pseudo-labelling for imbalanced SSL. We propose to learn refinement and thresholding parameters from a partition of the training dataset in a class-balanced way. SEVAL adapts to specific tasks with improved pseudo-labels accuracy and ensures pseudo-labels correctness on a per-class basis. Our experiments show that SEVAL surpasses state-of-the-art SSL methods, delivering more accurate and effective pseudo-labels in various imbalanced SSL situations. SEVAL, with its simplicity and flexibility, can enhance various SSL techniques effectively. The code is publicly available. Copyright © 2024, The Authors. All rights reserved.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Characterising directed and undirected metrics of high-order interdependence

arXiv

引用

arXiv 2024年

作者： Rosas, Fernando E. Mediano, Pedro A.M. Gastpar, Michael Department of Informatics University of Sussex BrightonBN1 9QJ United Kingdom Department of Computing Imperial College London LondonSW7 2RH United Kingdom School of Computer and Communication Sciences École Polytechnique Fédérale de Lausanne Lausanne1015 Switzerland

Systems of interest for theoretical or experimental work often exhibit high-order interactions, corresponding to statistical interdependencies in groups of variables that cannot be reduced to dependencies in subsets of them. While still under active development, the framework of partial information decomposition (PID) has emerged as the dominant approach to conceptualise and calculate high-order interdependencies. PID approaches can be grouped in two types: directed approaches that divide variables into sources and targets, and undirected approaches that treat all variables equally. Directed and undirected approaches are usually employed to investigate different scenarios, and hence little is known about how these two types of approaches may relate to each other, or if their corresponding quantities are linked in some way. In this paper we investigate the relationship between the redundancy-synergy index (RSI) and the O-information, which are practical metrics of directed and undirected high-order interdependencies, respectively. Our results reveal tight links between these two quantities, and provide two interpretations of them in terms of likelihood ratios in a hypothesis testing setting, as well as in terms of projections in information geometry. Copyright © 2024, The Authors. All rights reserved.

关键词： Well testing

来源：评论

学校读者我要写书评

暂无评论

Clarifying Assumptions about Artificial Intelligence before Revolutionising Patent Law

引用

GRUR International 2022年第4期71卷 295-321页

作者： Kim, Daria Alber, Maximilian Kwok, Man Wai Mitrovic, Jelena Ramirez-Atencia, Cristian Perez, Jesus Alberto Rodriguez Zille, Heiner Max Planck Institute for Innovation and Competition Munich Germany Charite Institute of Pathology Berlin Germany Chinese University of Hong Kong Department of Physics (Formerly) Hong Kong BMBF Program for Women in AI Faculty of Computer Science and Mathematics Chair of Data Science University of Passau Germany Department of Computer Systems Universidad Politecnica de Madrid Spain School of Computing Science College of Science and Engineering University of Glasgow United Kingdom Faculty of Computer Science Otto von Guericke University Magdeburg Germany

This paper examines several widespread assumptions about artificial intelligence, particularly machine learning, that are often taken as factual premises in discussions on the future of patent law in the wake of 'artificial ingenuity'. The objective is to draw a more realistic and nuanced picture of the human-computer interaction in solving technical problems than where 'intelligent' systems autonomously yield inventions. A detailed technical perspective is presented for each assumption, followed by a discussion of pertinent uncertainties for patent law. Overall, it is argued that implications of machine learning for the patent system in its core tenets appear far less revolutionary than is often posited. © 2022 Published by OUP and CH Beck on behalf of GRUR e.V.

关键词：

来源：评论

学校读者我要写书评

暂无评论

The Role of Internet of Things (IoT) to Fight against Covid-19

The Role of Internet of Things (IoT) to Fight against Covid-...

引用

Proceedings of the International Conference on Data Science, Machine Learning and Artificial Intelligence

作者： Md Alimul Haque Deepa Sonal Shameemul Haque Kailash Kumar Moidur Rahman Department of Computer Science Veer Kunwar Singh University Ara Bihar India Department of Computer Science Patna Women's College Patna Bihar India Al - Hafeez College Ara Bihar India College of Computing and Informatics Saudi Electronic University Riyadh Kingdom of Saudi Arabia College of Computing and Information Technology Jazan University Jizan Kingdom of Saudi Arabia

ISBN: (纸本)9781450387637

The Internet of Things (IoT) has become a new topic of research in various scientific and corporate fields, particularly in healthcare, in recent decades. The IoT revolution transforms new healthcare services by the inclusion of scientific, political and environmental prospects. It progresses from traditional to specialized healthcare systems, making it easier to identify, treat and manage patients. COVID - 19 has developed into a pandemic that spread around the globe. Day and night, scientists and engineers are working to produce a vaccine, develop additional research equipment and improve surveillance. Using contactless equipment is one of the critical factors to reduce the spread. To prevent the transmission of the coronavirus IoT can be used. That is an interconnection between the Internet and electronic devices. Devices are sensuous and recordable as well as can watch and respond. In this article, we discussed Covid-19 literature, monitoring techniques and proposed an IoT model to help mitigate the transmission of Covid-19.

关键词： COVID-19 Healthcare Internet of Things (IoT) Monitoring

来源：评论

学校读者我要写书评

暂无评论

Self-attention transfer networks for speech emotion recognition

引用

Virtual Reality & Intelligent Hardware 2021年第1期3卷 43-54页

作者： Ziping ZHAO Keru Wang Zhongtian BAO Zixing ZHANG Nicholas CUMMINS Shihuang SUN Haishuai WANG Jianhua TAO Björn WSCHULLER College of Computer and Information Engineering Tianjin Normal UniversityTianjin 300387China GLAM-Group on Language Audio&MusicImperial College LondonSW72AZUK Chair of Embedded Intelligence for Health Care and Wellbeing University of Augsburg86159Germany Department of Biostatistics and Health Informatics IoPPNKing's College LondonLondonSE58AFUK Department of Computer Science and Engineering Fairfield University 06824USA National Laboratory of Pattern Recognition CASIABeijing 100190China

Background A crucial element of human-machine interaction,the automatic detection of emotional states from human speech has long been regarded as a challenging task for machine learning *** vital challenge in speech emotion recognition(SER)is learning robust and discriminative representations from *** machine learning methods have been widely applied in SER research,the inadequate amount of available annotated data has become a bottleneck impeding the extended application of such techniques(e.g.,deep neural networks).To address this issue,we present a deep learning method that combines knowledge transfer and self-attention for SER ***,we apply the log-Mel spectrogram with deltas and delta-deltas as ***,given that emotions are time dependent,we apply temporal convolutional neural networks to model the variations in *** further introduce an attention transfer mechanism,which is based on a self-attention algorithm to learn long-term *** self-attention transfer network(SATN)in our proposed approach takes advantage of attention transfer to learn attention from speech recognition,followed by transferring this knowledge into *** evaluation built on Interactive Emotional Dyadic Motion Capture(IEMOCAP)dataset demonstrates the effectiveness of the proposed model.

关键词： Speech emotion recognition Attention transfer Self-attention Temporal convolutional neural networks(TCNs)

来源：评论

学校读者我要写书评

暂无评论

Deep Learning and AI for Superior Colorectal Polyp Detection with YOLOv9 Variants

Deep Learning and AI for Superior Colorectal Polyp Detection...

引用

Conference on Data Science and Machine Learning Applications (CDMA)

作者： Zeeshan Haider Ahmad Taher Azar Suliman Mohamed Fati Ibraheem Kasim Ibraheem Automated Systems and Soft Computing Lab (ASSCL) Prince Sultan University Riyadh Saudi Arabia College of Computer and Information Sciences Prince Sultan University Riyadh Saudi Arabia Faculty of Computers and Artificial Intelligence Benha University Benha Egypt Ibraheem Kasim Ibraheem is with Department of Electrical Engineering College of Engineering University of Baghdad Baghdad Iraq

ISBN: (数字)9798331539696

ISBN: (纸本)9798331539702

The growing interest in artificial intelligence (AI) as a diagnostic aid has spurred significant research into deep learning-based colorectal cancer polyp detection and localization. The motivation behind this research stems from the occasional missed diagnosis of polyps by medical experts due to fatigue or inexperience, which can lead to serious complications like colorectal cancer. In this research, we explore the effect on the accuracy of polyp detection and localization, employing variants of the state-of-the-art detection model YOLOv9. Four variants of YOLOv9 (gelan-e, gelan-c, yolov9-c, and yolov9-e) are trained on two datasets: CVC-CLinicDB and Kvasir combined as one dataset and LDPolyp Video as the second separate dataset. Results show that the YOLOv9 variants show their superiority when evaluated on five test datasets as compared to YOLOv8, the most recent YOLO series model.

关键词： Location awareness YOLO Deep learning Analytical models Accuracy Colorectal cancer Data science Fatigue Multiaccess communication Medical diagnostic imaging

来源：评论

学校读者我要写书评

暂无评论

Influenza Flu Diagnostics and Detection using Artificial Intelligence-based Learning Approaches: Challenges and Recent Study

Influenza Flu Diagnostics and Detection using Artificial Int...

引用

International Conference on Contemporary computing and informatics (IC3I)

作者： Yogesh Kumar G. Prabhu Kanna S J K Jagadeesh Kumar G. Sambasivam Department of Computer Science & Engineering Pandit Deendayal Energy University Gandhinagar Gujarat India School of Computer Science & Engineering VIT Bhopal University Bhopal Madhya Pradesh India Kathir College of Engineering Coimbatore Tamil Nadu India School of Computing and Data Science Xiamen University Malaysia Malaysia

Influenza is a life-threatening disease mainly for an older population. Common causes of death in Influenza related complications are pneumonia, obstructive pulmonary disease, cardiorespiratory diseases, ischemic heart, and pneumonia. In the past few decades, the death count has increased due to Influenza. So, in this paper, we have tried to give brief detail about various types of Influenza viruses forwarded with different methods that can be used for detecting them. Along with this, a literature survey of varying work done in the field of detecting Influenza using Machine learning, Deep learning, and Artificial intelligence is covered. In other section of the paper, we have given comparative study of the same work done in section 3 that helps researchers in getting brief knowledge about available datasets, Artificial intelligence algorithms is used with their outcomes and limitations. The motive of overall study is giving brief knowledge about Influenza flu and role of Artificial intelligence in it.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Advanced Machine Learning Algorithms for 1Predictive Analytics in Healthcare to Enhance Patient Outcomes with Data-Driven Insights

Advanced Machine Learning Algorithms for 1Predictive Analyti...

引用

Recent Advances in Science and engineering Technology (ICRASET), International Conference on

作者： Rajkumar Pandiarajan J. Jagannathan P.S. Ramesh A. Ponmalar Sudha Department of Information and Technology School of Computing Kalasalingam Academy Of Research and Education Krishnankoil Virudhunagar India Department of Software and Systems Engineering School of Computer Science Engineering and Information Systems Vellore Institute of Technology Katpadi Vellore India Department of Computer Science and Engineering Vel Tech Rangarajan Dr.Sagunthala R&D Institute of Science and Technology Chennai Tamilnadu India R.M.K. Engineering College Chennai Tamilnadu India Department of Computer Science and Engineering Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences SIMATS University Chennai Tamil Nadu India

ISBN: (数字)9798350388602

ISBN: (纸本)9798350388619

This research aims to explore the use of modern complex defensive machine learning algorithms in the provision of predictive analytics for health improvement. Incorporating electronic health records, medical image information, and genomic data the work proposes and assesses multiple machine learning models including deep learning, ensemble learning, and reinforcement learning. The main findings shown very high performance in disease prediction, in chronic illness and in the application of efficiency improvements in hospitals as a whole. Overall, such threats as data quality, data privacy, and model interpretability have been discussed in the research and emphasize the importance of the machine learning approach in healthcare. It is highlighted that further development and integration of the interdisciplinary approaches as well as the use of the transparent and interpretable artificial intelligence and machine learning models can be urgent to increase the acceptance among the healthcare staff. This paper enhances the literature on applying machine learning in healthcare systems adding to the repertoire of knowledge for improved patient care.

关键词： Data privacy Machine learning algorithms Hospitals Data integrity Machine learning Reinforcement learning Predictive models Data models Predictive analytics Diseases

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Driven - Pest Classification and Sustainable Agriculture

Deep Learning Driven - Pest Classification and Sustainable A...

引用

Electronics and Renewable Systems (ICEARS), International Conference on

作者： M Sirish Kumar R Usha B Rama Kantha Reddy S Ramachandra P Jaya Prakash B Swetha Dept. of Data Science School of Computing Mohan Babu University India CSE Madanapalle Institute of Technology and Science Madanapalle CSE (AI and ML) Sri Venkateswara College of Engineering Tirupati CSE Mother Theresa Institute of Engineering and Technology Palamaner Department of IT Sri Venkateswara College of Engineering Tirupati Dept. of Master of Computer Applications Mohan Babu University India

ISBN: (数字)9798331509675

ISBN: (纸本)9798331509682

This study introduces a system designed to identify pests in crops and classify them as either beneficial or harmful. The project begins by providing a comprehensive overview of various pest identification methods, analyzing their strengths and limitations. To address the problem, a categorization approach is proposed that incorporates convolutional neural networks (CNN) to build the pest identification and classification model. The model was trained using a dataset containing images of different pest species. Extensive testing was conducted on a large dataset, and the system's performance was validated against multiple traditional classification models. The proposed framework achieved a classification accuracy of 90%, outperforming conventional methods. This research highlights agriculture's vital role in the economy and underscores the potential of advanced technologies in supporting sustainable farming practices.

关键词： Deep learning Accuracy Insects System performance Transfer learning Crops Pesticides Agriculture Convolutional neural networks Testing

来源：评论

学校读者我要写书评

暂无评论

Spatiotemporal correlation based self-adaptive pose estimation in complex scenes

引用

Digital Communications and Networks 2024年

作者： Fu, Weina Luo, Zhe Liu, Shuai Lloret, Jaime de Albuquerque, Victor Hugo C. Saudagar, Abdul Khader Jilani Muhammad, Khan College of Information Science and Engineering Hunan Normal University Changsha410081 China School of Educational Science Hunan Normal University Changsha410081 China Universitat Politécnica de Valencia Gandia València46370 Spain Department of Teleinformatics Engineering Federal University of Ceará Ceará Fortaleza Brazil Riyadh11432 Saudi Arabia Department of Applied AI School of Convergence College of Computing and Informatics Sungkyunkwan University Seoul03063 Korea Republic of

In the current digital era, large-scale Artificial Intelligence (AI) models have fundamentally changed the landscape of the AI field. These models excel at extracting complex patterns and information from massive datasets, making them indispensable in a wide range of domains. With the advancement of large-scale AI models, pose estimation has achieved improved performance in managing large data sets and complex computations. However, most existing pose estimation methods suffer from the problem of inaccurate pose graph extraction in complex scenes. Moreover, the existing methods all focus on the message, ignoring the fact that it may not be optimal for the graph structure. This fixed graph structure is difficult to reflect the real dependency relationship between joint points. This study proposes a Character Redetection Model (CRM) to dynamically adjust the foreground extraction strategies to solve these problems. The CRM locates the target position based on the rich spatiotemporal information between the adjacent frames, effectively solving the problem of inaccurate foreground acquisition caused by occlusion and blur in complex scenes. In addition, an Uncertainty Graph Structure (UGS) is designed, which adaptively learns the topological structure of graphs with different Graph Convolutional Network (GCN) layers and skeleton samples in an end-to-end manner and mines the hidden space dependency between data to accurately model the different poses of the characters. The experimental results show that the proposed method achieves a superior performance compared to that of the State-Of-The-Art (SOTA) method, with a Percentage of Correct Keypoints (PCK) value of 94.6% at the Max Planck Institute for informatics (MPII), along with the mean Average Precision (mAP) values of 83.3% and 82.1% on the PoseTrack2017 and classroom pose datasets, respectively. © 2024 Chongqing University of Posts and Telecommunications

关键词： Network theory (graphs)

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：