检索结果-内蒙古大学图书馆

TEVA: Training-Efficient and Verifiable Aggregation for Federated Learning for Consumer Electronics in Industry 5.0

IEEE Transactions on Consumer Electronics 2024年

作者： Xia, Yuanjun Liu, Yining Chen, Jingxue Liang, Yangfan Khan, Fazlullah Alturki, Ryan Wang, Xiaopei Guilin University of Electronic Technology Guangxi Key Laboratory of Trusted Software School of Computer and Information Security Guilin541004 China Wenzhou University of Technology School of Data Science and Artificial Intelligence Wenzhou325027 China University of Technology School of Mathematical and Physical Sciences SydneyNSW2007 Australia Jiaxing University Provincial Key Laboratory of Multimodal Perceiving and Intelligent Systems The Key Laboratory of Medical Electronics and Digital Health of Zhejiang Province The Engineering Research Center of Intelligent Human Health Situation Awareness of Zhejiang Province Jiaxing314001 China University of Nottingham Ningbo China School of Computer Science Faculty of Science and Engineering Zhejiang Ningbo315104 China Umm AI-Qura University Makkah Department of Software Engineering College of Computing Saudi Arabia Riverside Department of Computer Science and Engineering University of California CA92521 United States

Federated learning (FL) has been widely used for privacy-preserving model updates in Industry 5.0, facilitated by 6G networks. Despite FL's privacy-preserving advantages, it remains vulnerable to attacks where adversaries can infer private data from local models or manipulate the central server (CS) to deliver falsified global models. Current privacy-preserving approaches, primarily based on the FedAvg algorithm, fail to optimize training efficiency for non-independent and identically distributed (non-IID) data. This article proposes training-efficient and verifiable aggregation (TEVA) for FL to resolve these issues. This scheme combines threshold Paillier homomorphic encryption (TPHE), verifiable aggregation, and an optimized double momentum update mechanism (OdMum). TEVA not only leverages TPHE to protect the privacy of local models but also ensures the integrity of the global model through a verifiable aggregation mechanism. Additionally, TEVA integrates the OdMum algorithm to effectively address the challenges posed by non-IID data, promoting rapid model convergence and significantly enhancing overall training efficiency. Security analysis indicates that TEVA meets the requirements for privacy protection. Extensive experimental results demonstrate that TEVA can accelerate model convergence while incurring lower computational and communication overheads. © 1975-2011 IEEE.

关键词： Differential privacy

来源：评论

学校读者我要写书评

暂无评论

Building human Values into Recommender Systems: An Interdisciplinary Synthesis

引用

ACM Transactions on Recommender Systems 2024年第3期2卷 1-57页

作者： Jonathan Stray Alon Halevy Parisa Assar Dylan Hadfield-Menell Craig Boutilier Amar Ashar Chloe Bakalar Lex Beattie Michael Ekstrand Claire Leibowicz Connie Moon Sehat Sara Johansen Lianne Kerlin David Vickrey Spandana Singh Sanne Vrijenhoek Amy Zhang McKane Andrus Natali Helberger Polina Proutskova Tanushree Mitra Nina Vasan UC Berkeley Center for Human-compatible AI Berkeley USA Meta AI Menlo Park USA Adobe Inc. San Jose USA Department of Electrical Engineering and Computer Science MIT Cambridge USA Google Research Mountain View USA Spotify Inc. New York City USA Meta Inc. Menlo Park USA Department of Information Science Drexel University Philadelphia USA Partnership on AI Palo Alto USA Hacks/Hackers Atlanta USA Department of Psychiatry and Behavioral Sciences Stanford University Stanford USA BBC London UK Institute for Information Law University of Amsterdam Amsterdam Netherlands Allen School of Computer Science and Engineering University of Washington Seattle USA Information School University of Washington Seattle USA Department of Psychiatry and Behavioral Sciences Stanford University Palo Alto USA

Recommender systems are the algorithms which select, filter, and personalize content across many of the world's largest platforms and apps. As such, their positive and negative effects on individuals and on societies have been extensively theorized and studied. Our overarching question is how to ensure that recommender systems enact the values of the individuals and societies that they serve. Addressing this question in a principled fashion requires technical knowledge of recommender design and operation, and also critically depends on insights from diverse fields including social science, ethics, economics, psychology, policy, and law. This article is a multidisciplinary effort to synthesize theory and practice from different perspectives, with the goal of providing a shared language, articulating current design approaches, and identifying open problems. We collect a set of values that seem most relevant to recommender systems operating across different domains, and then examine them from the perspectives of current industry practice, measurement, product design, and policy approaches. Important open problems include multi-stakeholder processes for defining values and resolving trade-offs, better values-driven measurements, recommender controls that people use, non-behavioral algorithmic feedback, optimization for long-term outcomes, causal inference of recommender effects, academic-industry research collaborations, and interdisciplinary policy-making.

关键词： Recommender systems value-sensitive design technology policy

来源：评论

学校读者我要写书评

暂无评论

Evaluation of Different Radar Placements for Food Intake Monitoring Using Deep Learning

Evaluation of Different Radar Placements for Food Intake Mon...

引用

IEEE National Conference on Radar

作者： Chunzhuo Wang T. Sunil Kumar Walter De Raedt Guido Camps Hans Hallez Bart Vanrumste The e-Media Research Lab and the ESAT-STADIUS Division KU Leuven Leuven Belgium The Life Science Department IMEC Heverlee Belgium Department of Agrotechnology and Food Sciences The Division of Human Nutrition and Health Wageningen University and Research and the OnePlanet Research Center Wageningen Netherlands Department of Computer Science The M-Group DistriNet KU Leuven Sint-Michiels Belgium

Automated food intake monitoring has drawn significant attention due to its potential applications in the healthcare domain. Plenty of research, including wrist-worn imu-based and camera-based approaches, have emerged to detect food intake activity passively and objectively. Recently, researchers explored radar for food intake monitoring because of its contactless and privacy-preserving characteristics. In this study, we deploy the Frequency Modulated Continuous Wave (FMCW) radar in three different positions to investigate the performance of each position in automated eating gesture detection. The three positions are front, side, and overhead. Fifteen participants are recruited to have three meals (45 meals, 641 min in total), while the radar is deployed in different positions in each meal. A 3D Temporal Convolutional Network (3D-TCN) is used to process the range-doppler cube (RD Cube) of each dataset. The Leave-One-Subject-Out (LOSO) validation method shows that putting radar in the front position obtains the best performance with a segmental F1-score of 0.786 and 0.825 for eating and drinking gestures, respectively.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Survey and Improvement Strategies for Gene Prioritization with Large Language Models

arXiv

引用

arXiv 2025年

作者： Neeley, Matthew B. Qi, Guantong Wang, Guanchu Tang, Ruixiang Mao, Dongxue Liu, Chaozhong Pasupuleti, Sasidhar Yuan, Bo Xia, Fan Liu, Pengfei Liu, Zhandong Hu, Xia Jan and Dan Duncan Neurological Research Institute Texas Children’s Hospital Houston United States Graduate School of Biomedical Sciences Program in Quantitative and Computational Biosciences Baylor College of Medicine Houston United States Graduate School of Biomedical Sciences Program in Genetics and Genomics Baylor College of Medicine Houston United States Department of Computer Science Rice University Houston United States Department of Computer Science Rutgers University NJ United States Department of Pediatrics Baylor College of Medicine Houston United States Department of Molecular and Human Genetics Baylor College of Medicine Houston United States Human Genome Sequencing Center Baylor College of Medicine Houston United States Baylor Genetics Houston United States

Background Rare diseases can be difficult to diagnose due to limited patient data and broad genetic diversity. Despite the advances in variant prioritization tools, many rare disease cases remain undiagnosed. While large language models (LLMs) have performed well in medical exams, their accuracy in diagnosing rare genetic diseases has not yet been evaluated. Methods To identify causal genes for genetic diseases, we benchmarked various LLMs for gene prioritization. We employed multi-agent and human Phenotype Ontology (HPO) classification approaches to identify patient case groups based on phenotypes categorizing them into different levels of solvability. To address LLM limitations in ranking large gene sets, we used a divide-and-conquer strategy to break down the ranking task into smaller subsets. Mini-batching inputs and limiting the number of generated tokens improved efficiency. Results In its vanilla form, GPT-4 consistently outperformed the other LLMs with an accuracy around 30%. Multi-agent and HPO classification approaches aided us in distinguishing between confidently-solved and challenging cases. In addition, we observed bias towards well-studied genes and input order sensitivity which are drawbacks of LLMs in disease-causal gene prioritization. Our divide-and-conquer strategy enhanced accuracy by overcoming positional and gene frequency biases in literature. This framework, based on benchmarking insights and novel techniques, significantly optimizes the overall process when identifying disease-causal genes for genetic diseases when compared to our baseline evaluation, and thereby better enables the development of targeted diagnostic and therapeutic interventions. Conclusions Using HPO classification and our novel multi-agent techniques, and our LLM divide-and-conquer strategy (1) highlighted the importance of accounting for differences in patient case solvability and (2) yielded improved performance in identifying causal genes for rare diseases when compare

关键词： Gene Ontology

来源：评论

学校读者我要写书评

暂无评论

Image-based Gait Spatiotemporal Parameters Estimation using a Single Camera and CNN-Transformer Hybrid Network

Image-based Gait Spatiotemporal Parameters Estimation using ...

引用

Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)

作者： Ankhzaya Jamsrandorj Quynh Hoang Ngan Nguyen Dawoon Jung Min Seok Baek Kyung-Ryoul Mun Jinwook Kim Department of Human Computer Interface & Robotics Engineering KIST School University of Science & Technology Seoul Republic of Korea Center for Artificial Intelligence Korea Institute of Science and Technology Seoul Republic of Korea Department of AI Robotics KIST School University of Science & Technology Seoul Republic of Korea Department of Neurology Wonju Severance Christian Hospital Yonsei University Wonju College of Medicine Wonju Republic of Korea

Vision-based gait analysis can play an important role in the remote and continuous monitoring of the elderly’s health conditions. However, most vision-based approaches compute gait spatiotemporal parameters using human pose information and provide average parameters. This study aimed to propose a straightforward method for stride-by-stride gait spatiotemporal parameters estimation. A total of 160 elderly individuals participated in this study. Data were gathered with a GAITRite system and a mobile camera simultaneously. Three deep learning networks were trained with a few RGB frames as input and a continuous 1D signal containing both spatial and temporal gait parameters as output. The trained networks estimated the stride lengths with correlations of 0.938 and more and detected gait events with F 1 -scores of 0.914 and *** relevance— The proposed method showed excellent agreements with the GAITRite system in analyzing spatiotemporal gait parameters. Our approach can be applied to monitor the elderly’s health conditions based on their gait parameters for early diagnosis of diseases, proper treatment, and timely intervention.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Intake Gesture Detection With IMU Sensor in Free-Living Environments: The Effects of Measuring Two-Hand Intake and Down-Sampling

Intake Gesture Detection With IMU Sensor in Free-Living Envi...

引用

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

作者： Chunzhuo Wang Jiaze Kong Yutong Cai T. Sunil Kumar Walter De Raedt Guido Camps Hans Hallez Bart Vanrumste The e-Media Research Lab and the ESAT-STADIUS Division KU Leuven Leuven Belgium The Life Science Department IMEC Heverlee Belgium Department of Agrotechnology and Food Sciences The Division of Human Nutrition and Health Wageningen University and Research and the OnePlanet Research Center Wageningen The Netherlands Department of Computer Science The M-Group DistriNet KU Leuven Sint-Michiels Belgium

Food intake monitoring plays an important role in personal dietary systems. Numerous approaches have been proposed to automatically detect eating gestures using various sensors and machine learning. However, existing eating gesture detection approaches mainly focus on meal sessions. Such a task is still challenging in free-living environments due to longer monitoring duration and more non-feeding activities. This paper proposes a wearable Inertial Measurement Unit (IMU) based method to detect eating and drinking gestures in free-living environments. Two important factors that impede intake gesture detection in free-living environments are addressed: 1) how to handle IMU data from two hands, and 2) what is the impact of downsampling sensor data on performance. To integrate two-hand data, we propose a solution that combines hand mirroring and temporal concatenation techniques. The multi-stage temporal convolutional network (MS-TCN) is applied to effectively recognise intake gestures. A dataset contains 12 subjects with 67.5 h data is collected for validation. Moreover, IMU data with different sampling frequencies are processed to test performance. Validated by Leave-One-Subject-Out (LOSO) method, our approach (with 16 Hz sampling frequency) achieves a segmental F1-score of 0.826 and 0.893 for recognizing eating and drinking gestures, respectively. Results show that the proposed solution outperforms existing two-hand data combination approaches. Moreover, in our case, a higher sampling frequency does not always mean better performance.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Explaining the Black-box Smoothly - A Counterfactual Approach

arXiv

引用

arXiv 2021年

作者： Singla, Sumedha Eslami, Motahhare Pollack, Brian Wallace, Stephen Batmanghelich, Kayhan Computer Science Department The University of Pittsburgh PittsburghPA15206 United States School of Computer Science Human-Computer Interaction Institute Carnegie Mellon University United States Department of Biomedical Informatics The University of Pittsburgh PittsburghPA15206 United States University of Pittsburgh Medical School PittsburghPA15206 United States

We propose a BlackBox Counterfactual Explainer, designed to explain image classification models for medical applications. Classical approaches (e.g., saliency maps) that assess feature importance do not explain how imaging features in important anatomical regions are relevant to the classification decision. Such reasoning is crucial for transparent decision-making in healthcare applications. Our framework explains the decision for a target class by gradually exaggerating the semantic effect of the class in a query image. We adopted a Generative Adversarial Network (GAN) to generate a progressive set of perturbations to a query image, such that the classification decision changes from its original class to its negation. Our proposed loss function preserves essential details (e.g., support devices) in the generated images. We used counterfactual explanations from our framework to audit a classifier trained on a chest x-ray dataset with multiple labels. Clinical evaluation of model explanations is a challenging task. We proposed clinically-relevant quantitative metrics such as cardiothoracic ratio and the score of a healthy costophrenic recess to evaluate our explanations. We used these metrics to quantify the counterfactual changes between the populations with negative and positive decisions for a diagnosis by the given classifier. We conducted a human-grounded experiment with diagnostic radiology residents to compare different styles of explanations (no explanation, saliency map, cycleGAN explanation, and our counterfactual explanation) by evaluating different aspects of explanations: (1) understandability, (2) classifier's decision justification, (3) visual quality, (d) identity preservation, and (5) overall helpfulness of an explanation to the users. Our results show that our counterfactual explanation was the only explanation method that significantly improved the users' understanding of the classifier's decision compared to the no-explanation baseline. Our metric

关键词： Medical applications

来源：评论

学校读者我要写书评

暂无评论

Investigating the interaction Experience with Augmented Reality Based Writing Tools 21st

Investigating the Interaction Experience with Augmented Real...

引用

21st International Conference on human-computer interaction, HCI International 2019

作者： Zheng, Wei Yi, Xin Shi, Yuanchun Pervasive Human-Computer Interaction Group Department of Computer Science and Technology Tsinghua University Beijing100084 China

ISBN: (纸本)9783030235277

The contents of this paper include the design methodology of perceptual computing products, the innovative design of the AR writing and its technical feasibility, the writing interactive experience test of different hardware platforms, etc. It has certain reference value for the research work of paperless writing. © Springer Nature Switzerland AG 2019.

关键词： Augmented reality

来源：评论

学校读者我要写书评

暂无评论

Two Faces of Privacy: Legal and human-centered Perspectives of Lifelogging Applications in Home Environments 6th

Two Faces of Privacy: Legal and Human-Centered Perspectives ...

引用

6th International Conference on human Aspects of IT for the Aged Population, ITAP 2020, held as part of the 22nd International Conference on human-computer interaction, HCII 2020

作者： Wilkowska, Wiktoria Offermann-van Heek, Julia Colonna, Liane Ziefle, Martina Human-Computer Interaction Center RWTH Aachen University Campus-Boulevard 57 Aachen52074 Germany Department of Law Stockholm University Universitetvägen 10C Stockholm10691 Sweden

ISBN: (纸本)9783030502485

In view of the consequences resulting from the demographic change, using assisting lifelogging technologies in domestic environments represents one potential approach to support elderly and people in need of care to stay longer within their own home. Yet, the handling of personal data poses a considerable challenge to the perceptions of privacy and data security, and therefore for an accepted use in this regard. The present study focuses on aspects of data management in the context of two different lifelogging applications, considering a legal and a human-centered perspective. In a two-step empirical process, consisting of qualitative interviews and an online survey, these aspects were explored and evaluated by a representative German sample of adult participants (N = 209). Findings show positive attitudes towards using lifelogging, but there are also high requirements on privacy and data security as well as anonymization of the data. In addition, the study allows deep insights into preferred duration and location of the data storage, and permissions to access the personal information from third parties. Knowledge of preferences and requirements in the area of data management from the legal and human-centered perspectives is crucial for lifelogging and must be considered in applications that support people in their daily living at home. Outcomes of the present study considerably contribute to the understanding of an optimal infrastructure of the accepted and willingly utilized lifelogging applications. © 2020, Springer Nature Switzerland AG.

关键词： Information management

来源：评论

学校读者我要写书评

暂无评论

Evoking Stress Reactivity in a Virtual Dance Competition

Evoking Stress Reactivity in a Virtual Dance Competition

引用

AHFE Virtual Conference on human Factors and Simulation and the International Conference on Digital human Modeling and Applied Optimization, 2020

作者： van Dammen, Lotte Barnett, Neil Conrady, Roselynn Wright, Lucas Thymes, Bradon Shirtcliff, Elizabeth A. Department of Human Development and Family Studies Iowa State University AmesIA United States Virtual Reality Application Center Iowa State University AmesIA United States Department of Mechanical Engineering San Diego State University San DiegoCA United States Hamilton College Physics Department Hamilton ClintonNY United States Computer Science Department Hamilton ClintonNY United States Electrical Engineering and Computer Science Department Howard University WashingtonDC United States

ISBN: (纸本)9783030510633

Stress reactivity involves a physiological response to a stressful task, as a biological mechanism related to mental and physical health outcomes. Traditional stressors evoke social evaluative threat, which is fear of judgment and a threat to social status, typically via public speaking tasks. To be well-suited for virtual reality, however, a novel approach is necessary in the design of new stress tasks. We hypothesized that a virtual reality stress task involving social evaluative threat through a dance competition in front of an audience, elicits a stress response, measured by autonomic nervous system, cortisol, and testosterone reactivity. Participants (n = 18) showed autonomic nervous system reactivity in terms of increased heart rate and decreased respiratory sinus arrhythmia, indicating a stress response. Levels of cortisol increased in response to the dance competition, especially within responders, whereas testosterone levels did not change significantly over time. A virtual reality dance competition involving physical social evaluative threat elicits a stress response. © 2021, The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Virtual reality

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：