检索结果-内蒙古大学图书馆

Fair Evaluation of Federated Learning Algorithms for Automated Breast Density Classification: The Results of the 2022 ACR-NCI-NVIDIA Federated Learning Challenge

arXiv

引用

arXiv 2024年

作者： Schmidt, Kendall Bearce, Benjamin Chang, Ken Coombs, Laura Farahani, Keyvan Elbatel, Marawan Mouheb, Kaouther Marti, Robert Zhang, Ruipeng Zhang, Yao Wang, Yanfeng Hu, Yaojun Ying, Haochao Xu, Yuyang Testagrose, Conrad Demirer, Mutlu Gupta, Vikash Akünal, Ünal Bujotzek, Markus Maier-Hein, Klaus H. Qin, Yi Li, Xiaomeng Kalpathy-Cramer, Jayashree Roth, Holger R. American College of Radiology United States The Massachusetts General Hospital United States University of Colorado United States National Institutes of Health National Cancer Institute United States Computer Vision and Robotics Institute University of Girona Spain Cooperative Medianet Innovation Center Shanghai Jiao Tong University China Shanghai AI Laboratory China Real Doctor AI Research Centre Zhejiang University China School of Public Health Zhejiang University China College of Computer Science and Technology Zhejiang University China University of North Florida College of Computing Jacksonville United States Mayo Clinic Florida Radiology United States Division of Medical Image Computing German Cancer Research Center Heidelberg Germany Electronic and Computer Engineering Hong Kong University of Science and Technology China NVIDIA United States

The correct interpretation of breast density is important in the assessment of breast cancer risk. AI has been shown capable of accurately predicting breast density, however, due to the differences in imaging characteristics across mammography systems, models built using data from one system do not generalize well to other systems. Though federated learning (FL) has emerged as a way to improve the generalizability of AI without the need to share data, the best way to preserve features from all training data during FL is an active area of research. To explore FL methodology, the breast density classification FL challenge was hosted in partnership with the American College of Radiology, Harvard Medical Schools' Mass General Brigham, University of Colorado, NVIDIA, and the National Institutes of Health National Cancer Institute. Challenge participants were able to submit docker containers capable of implementing FL on three simulated medical facilities, each containing a unique large mammography dataset. The breast density FL challenge ran from June 15 to September 5, 2022, attracting seven finalists from around the world. The winning FL submission reached a linear kappa score of 0.653 on the challenge test data and 0.413 on an external testing dataset, scoring comparably to a model trained on the same data in a central location. © 2024, CC BY.

关键词： Learning algorithms

来源：评论

学校读者我要写书评

暂无评论

Towards Narrowing the Search in Bounded-Suboptimal Safe Interval Path Planning 14

Towards Narrowing the Search in Bounded-Suboptimal Safe Inte...

引用

14th International Symposium on Combinatorial Search, SoCS 2021

作者： Rybecky, Tomas Kulich, Miroslav Andreychuk, Anton Yakovlev, Konstantin Department of Cybernetics Faculty of Electrical Engineering Czech Technical University Czech Republic Czech Institute for Informatics Robotics and Cybernetics Czech Technical University Czech Republic Federal Research Center for Computer Science and Control of Russian Academy of Sciences Russia Russia National Research University Higher School of Economics Russia

ISBN: (纸本)9781713834557

Path planning in the presence of dynamic obstacles is challenging as the time dimension has to be considered. A prominent approach to tackle this problem known to be complete and optimal is the A*-based Safe-interval Path Planning (SIPP). Bounded-suboptimal variants of SIPP employing the ideas of Weighted A* (WSIPP) and Focal Search (FocalSIPP) have been introduced recently, trading-off optimality for decreased planning time. In this paper, we revisit FocalSIPP and design several secondary heuristics for Focal Search with the intention to narrow the search in the direction of a preplanned optimal single-agent path not considering dynamic obstacles. The experimental results on various maps show that the designed heuristics generally outperform the hops-to-the-goal heuristic used in the original FocalSIPP and successfully compete with WSIPP as well. Copyright © 2021, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Motion planning

来源：评论

学校读者我要写书评

暂无评论

Optimization Design for Federated Learning in Heterogeneous 6G Networks

arXiv

引用

arXiv 2023年

作者： Luo, Bing Ouyang, Xiaomin Sun, Peng Han, Pengchao Ding, Ningning Huang, Jianwei The Data Science Research Center the Division of Natural and Applied Sciences Duke Kunshan University Jiangsu Kunshan China The department of Information Engineering The Chinese University of Hong Kong Hong Kong The College of Computer Science and Electronic Engineering Hunan University Changsha410082 China The School of Science and Engineering The Chinese University of Hong Kong Shenzhen China Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen China The Department of Electrical and Computer Engineering Northwestern University EvanstonIL60208 United States

With the rapid advancement of 5G networks, billions of smart Internet of Things (IoT) devices along with an enormous amount of data are generated at the network edge. While still at an early age, it is expected that the evolving 6G network will adopt advanced artificial intelligence (AI) technologies to collect, transmit, and learn this valuable data for innovative applications and intelligent services. However, traditional machine learning (ML) approaches require centralizing the training data in the data center or cloud, raising serious user-privacy concerns. Federated learning, as an emerging distributed AI paradigm with privacy-preserving nature, is anticipated to be a key enabler for achieving ubiquitous AI in 6G networks. However, there are several system and statistical heterogeneity challenges for effective and efficient FL implementation in 6G networks. In this article, we investigate the optimization approaches that can effectively address the challenging heterogeneity issues from three aspects: incentive mechanism design, network resource management, and personalized model optimization. We also present some open problems and promising directions for future research. Copyright © 2023, The Authors. All rights reserved.

关键词： 5G mobile communication systems

来源：评论

学校读者我要写书评

暂无评论

Unlock Pose Diversity: Accurate and Efficient Implicit Keypoint-based Spatiotemporal Diffusion for Audio-driven Talking Portrait

arXiv

引用

arXiv 2025年

作者： Yang, Chaolong Yao, Kai Yan, Yuyao Jiang, Chenru Zhao, Weiguang Sun, Jie Cheng, Guangliang Zhang, Yifei Dong, Bin Huang, Kaizhu Department of Computer Science University of Liverpool LiverpoolL69 7ZX United Kingdom Ant Group Hangzhou310000 China School of Robotic Xi’an Jiaotong-Liverpool University Suzhou215123 China Department of Mechatronics and Robotics Xi’an Jiaotong-Liverpool University Suzhou215123 China Department of Foundational Mathematics Xi’an Jiaotong-Liverpool University Suzhou215123 China Digital Innovation Research Center Duke Kunshan University Kunshan215316 China Ricoh Software Research Center Beijing100027 China

Audio-driven single-image talking portrait generation plays a crucial role in virtual reality, digital human creation, and filmmaking. Existing approaches are generally categorized into keypoint-based and image-based methods. Keypoint-based methods effectively preserve character identity but struggle to capture fine facial details due to the fixed points limitation of the 3D Morphable Model. Moreover, traditional generative networks face challenges in establishing causality between audio and keypoints on limited datasets, resulting in low pose diversity. In contrast, image-based approaches produce high-quality portraits with diverse details using the diffusion network but incur identity distortion and expensive computational costs. In this work, we propose KDTalker, the first framework to combine unsupervised implicit 3D keypoint with a spatiotemporal diffusion model. Leveraging unsupervised implicit 3D keypoints, KDTalker adapts facial information densities, allowing the diffusion process to model diverse head poses and capture fine facial details flexibly. The custom-designed spatiotemporal attention mechanism ensures accurate lip synchronization, producing temporally consistent, high-quality animations while enhancing computational efficiency. Experimental results demonstrate that KDTalker achieves state-of-the-art performance regarding lip synchronization accuracy, head pose diversity, and execution efficiency. Our codes are available at https://***/chaolongy/KDTalker. © 2025, CC BY-NC-ND.

关键词： Virtual reality

来源：评论

学校读者我要写书评

暂无评论

A Survey of the Self Supervised Learning Mechanisms for vision Transformers

arXiv

引用

arXiv 2024年

作者： Khan, Asifullah Sohail, Anabia Fiaz, Mustansar Hassan, Mehdi Afridi, Tariq Habib Marwat, Sibghat Ullah Munir, Farzeen Ali, Safdar Naseem, Hannan Zaheer, Muhammad Zaigham Ali, Kamran Sultana, Tangina Tanoli, Ziaurrehman Akhter, Naeem Pattern Recognition Lab DCIS PIEAS Nilore Islamabad45650 Pakistan PIEAS Nilore Islamabad45650 Pakistan Deep Learning Lab Center for Mathematical Sciences PIEAS Nilore Islamabad45650 Pakistan Center of Secure Cyber-Physical Security Systems Khalifa University Abu Dhabi United Arab Emirates IBM Research United States Department of Computer Science Air University Islamabad Pakistan Department of Computer Science and Engineering Kyung Hee University Global Campus 1732 Gyeonggi-do Yongin17104 Korea Republic of Department of Electrical Engineering and Automation Aalto University Finland Finnish Center of Artificial Center Finland Faculty of Engineering and Green Technology Universiti Tunku Abdul Rahman Malaysia Computer Vision Department Mohamed Bin Zayed University of Artificial Intelligence United Arab Emirates Karachi Pakistan Department of Electronics and Communication Engineering Hajee Mohammad Danesh Science and Technology University Bangladesh HiLIFE University of Helsinki Finland

vision Transformers (ViTs) have recently demonstrated remarkable performance in computer vision tasks. However, their parameter-intensive nature and reliance on large amounts of data for effective performance have shifted the focus from traditional human-annotated labels to unsupervised learning and pretraining strategies that uncover hidden structures within the data. In response to this challenge, self-supervised learning (SSL) has emerged as a promising paradigm. SSL leverages inherent relationships within the data itself as a form of supervision, eliminating the need for manual labeling and offering a more scalable and resource-efficient alternative for model training. Given these advantages, it is imperative to explore the integration of SSL techniques with ViTs, particularly in scenarios with limited labeled data. Inspired by this evolving trend, this survey aims to systematically review SSL mechanisms tailored for ViTs. We propose a comprehensive taxonomy to classify SSL techniques based on their representations and pre-training tasks. Additionally, we discuss the motivations behind SSL, review prominent pre-training tasks, and highlight advancements and challenges in this field. Furthermore, we conduct a comparative analysis of various SSL methods designed for ViTs, evaluating their strengths, limitations, and applicability to different scenarios. 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

A Standard Interconnect Benchmark for a European In-orbit Services, Manufacturing and Assembly (ISMA) Demonstrator 73

A Standard Interconnect Benchmark for a European In-orbit Se...

引用

73rd International Astronautical Congress, IAC 2022

作者： Brinkmann, Wiebke Yüksel, Mehmed Jankovic, Marko Wirkus, Malte Saffer, Jona Soto, Isabel Gancet, Jeremi Letier, Pierre Schervan, Thomas A. Kreisel, Joerg Estable, Stephane Kirchner, Frank DFKI GmbH Robotics Innovation Center Robert-Hooke-Str.1 Bremen28359 Germany SENER Aeroespacial S.A Av. Zugazarte 56 Getxo48930 Spain Space Applications Services NV Leuvensesteenweg 325 Sint-Stevens-Woluwe1932 Belgium iBOSS GmbH Dennewartstrasse 25-27 Aachen52068 Germany Airbus Defence and Space GmbH Willy-Messerschmitt-Strasse 1 Taufkirchen82024 Germany Robotics Research Group Department of Mathematics and Computer Science University of Bremen Germany

Growing space debris is an issue for which solutions are being sought, especially with the usage of space robotics. The topic ranges from disposal to sustainability. Modular robotic can be seen is a key factor to support sustainability in space. Within this framework, it is possible to combine modular components in such a way that, for example, a satellite can be created or in the event of malfunction, modules can be replaced without having to abandon the whole satellite. This reduces space debris. To connect the modules, standard interconnects (SIs) with multifunctional features, like to connect mechanically and transmit power and data, are required. In the operational grant (OG) PERIOD of the EU Horizon 2020 project PERASPERA, three existing SIs have been evaluated within a benchmarking concept to give a recommendation on the most suited one to be used in the orbital demonstration mission of PERIOD, as well as provide feedback for their future improvements. Testing was conducted by the German research center for Artificial Intelligence GmbH (DFKI) as an independent body to neutrally evaluate the performance of SIs in relevant demonstration scenarios and in full transparency to consortium members. This paper describes the benchmark approach, methodology, test setup, execution, and recommendation path for what concerns the mechanical aspects of SIs. The approach can be extended and applied to future deployments of SIs in European and international space projects. Copyright © IAC 2022 by Wiebke Brinkmann. Published by the International Astronautical Federation (IAF), with permission and released to the IAF to publish in all forms.

关键词： Space debris

来源：评论

学校读者我要写书评

暂无评论

An In-Silico Technique: Prediction of Unknown Protein Structure from Sequences by Homology

An In-Silico Technique: Prediction of Unknown Protein Struct...

引用

Information Technology (ACIT)

作者： Wafa Faisal Hana Eltaveb Mohammed Hassan Nusiba Ibrahim Faculty of Computer Science and Information Technology Sudan University of Science and Technology Khartoum Sudan National University Biomedical Research Institute National University Khartoum Sudan Department of Biology and Bioinformatics' 7Vision Training and Development Center Khartoum Sudan

ISBN: (数字)9798350320244

ISBN: (纸本)9798350320251

The structural characterization of protein sequences is one of the most common problems in biology. This task is usually facilitated by the accurate Three-Dimensional (3-D) structure of the protein. This paper was carried out to investigate and compare the most suitable software online tools for homology modeling of protein structure by obtaining few samples of proteins. Different length, known and unknown (3-D) proteins structures of bacteria, and viruses were obtained from Uniprot which do not have a Protein Data Bank Identifier (PDB ID). This paper used several tools, such CPHmodels-3.2, Phyre2, Swiss- Model, (PS)2-v2, and Robetta, for predicting, comparison and model generation, where CPHmodels-3.2 and Phyre2 Server, gave relatively better values. Proteins (3-D) structure was evaluated using PROSESS Server and the better evaluation to proteins structure was for the Phyre2 software online tool.

关键词： Proteins Microorganisms Databases Crystals Predictive models Software Servers

来源：评论

学校读者我要写书评

暂无评论

Walking-in-Place Characteristics-Based Geriatric Assessment Using Deep Convolutional Neural Networks 42

Walking-in-Place Characteristics-Based Geriatric Assessment ...

引用

42nd Annual International Conferences of the IEEE Engineering in Medicine and Biology Society, EMBC 2020

作者： Jung, Dawoon Nguyen, Mau Dung Park, Mina Kim, Miji Won, Chang Won Kim, Jinwook Mun, Kyung-Ryoul Center for Imaging Media Research Kist Seoul Korea Republic of University of Science and Technology Department of Human Computer Interface Robotics Engineering Daejeon Korea Republic of

ISBN: (纸本)9781728119908

The world population is aging, and this phenomenon is expected to continue for the next decades. This study aimed to propose a simple and reliable method that can be used for daily in-home monitoring of frailty and cognitive dysfunction in the elderly based on their walking-in-place characteristics. Fifty-four community-dwelling elderly people aged 65 years or older participated in this study. The participants were categorized into the robust and the non-robust groups according to the FRAIL scale. The mini-mental state examination was used to classify the cognitive impairment and the non-cognitive impairment groups. The 3-axis acceleration and the 3-axis angular velocity signals were measured using the inertial measurement units attached to the foot, shank, thigh, and posterior pelvis while each participant was walking in place for 20 seconds. The walking-in-place spectrograms were acquired by applying time-frequency analysis to the lower body movement signals measured in one stride. Four-fold cross-validation was applied to 80% of the total samples and the remaining 20% were used as test data. The deep convolutional neural network-based classifiers trained with the walking-inplace spectrograms enabled to categorize the robust and the non-robust groups with 94.63% accuracy and classify the cognitive impairment and the non-cognitive impairment groups with 97.59% accuracy. This study suggests that the walking-in-place spectrograms, which can be obtained without spacious experimental space, cumbersome equipment, and laborious processes, are effective indicators of frailty and cognitive dysfunction in the elderly. © 2020 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

HerStory: combating gender stereotypes by highlighting women’s contribution to the history of societies

HerStory: combating gender stereotypes by highlighting women...

引用

International Symposium on computers in Education (SIIE)

作者： Lucía García-Holgado Alicia García-Holgado Sonia Verdugo-Castro Erika García Silva Francisco J. García-Peñalvo Antrea Kosta Dimitra Sofianou Emily Psara Miguel Á. Conde GRIAL Research Group Research Institute for Educational Sciences Universidad de Salamanca Salamanca Spain Advisory Center for the Support of the Family SIKESO Nicosia Cyprus KEAN-Cell of Alternative Youth Activities Athens Greece Center for Social Innovation Cyprus Nicosia Cyprus Department of Mechanics Computer Science and Aerospace Engineering Robotics Group Universidad de León León Spain

HerStory project addresses the need of European Union (EU) to combat gender-based stereotypes, a priority underscored by the EU Council. Also, it adheres to the main aim of the EU Gender Equality strategy 2020-2025, of "Leading equally throughout society" as well as the objectives of the Gender Action Plan III. The project’s goal is to highlight the significant contribution of women in history across European cities. HerStory aims to develop a transnational map that will feature important women from different regions shedding light on important historical moments where women’s actions were decisive. This initiative will tackle the gender stereotypes that often exclude women from decision-making bodies, whether in politics and social life or work. At the same time, it will promote positive role models operating as an empowerment tool for girls and women. The project’s outcomes will provide capacity building and useful methods to youth experts, educators, academics representing a conscious effort to shift gender-related attitudes through education and youth engagement.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Adaptive Wavelet Filters as Practical Texture Feature Amplifiers for Parkinson’s Disease Screening in OCT

arXiv

引用

arXiv 2025年

作者： Zhang, Xiaoqing Shi, Hanfeng Li, Xiangyu Ye, Haili Xu, Tao Li, Na Hu, Yan Lv, Fan Chen, Jiangfan Liu, Jiang Center for High Performance Computing and Shenzhen Key Laboratory of Intelligent Bioinformatics Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen China Research Institute of Trustworthy Autonomous Systems Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen518055 China State Key Laboratory of Ophthalmology Optometry and Vision Science Wenzhou Medical University Wenzhou China Centre for Computational Science and Mathematical Modelling Coventry University Coventry United Kingdom School of Computer Science University of Nottingham Ningbo China Ningbo315100 China School of Ophthalmology and Optometry Wenzhou Medical University Wenzhou325035 China Department of Electronic and Information Engineering Changchun University Changchun130022 China

Parkinson’s disease (PD) is a prevalent neurodegenerative disorder globally. The eye’s retina is an extension of the brain and has great potential in PD screening. Recent studies have suggested that texture features extracted from retinal layers can be adopted as biomarkers for PD diagnosis under optical coherence tomography (OCT) images. Frequency domain learning techniques can enhance the feature representations of deep neural networks (DNNs) by decomposing frequency components involving rich texture features. Additionally, previous works have not exploited texture features for automated PD screening in OCT. Motivated by the above analysis, we propose a novel Adaptive Wavelet Filter (AWF) that serves as the Practical Texture Feature Amplifier to fully leverage the merits of texture features to boost the PD screening performance of DNNs with the aid of frequency domain learning. Specifically, AWF first enhances texture feature representation diversities via channel mixer, then emphasizes informative texture feature representations with the well-designed adaptive wavelet filtering token mixer. By combining the AWFs with the DNN stem, AWFNet is constructed for automated PD screening. Additionally, we introduce a novel Balanced Confidence (BC) Loss by mining the potential of sample-wise predicted probabilities of all classes and class frequency prior, to further boost the PD screening performance and trustworthiness of AWFNet. The extensive experiments manifest the superiority of our AWFNet and BC over state-of-the-art methods in terms of PD screening performance and trustworthiness. © 2025, CC BY-NC-SA.

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：