Jove
Visualize
联系我们
JoVE
x logofacebook logolinkedin logoyoutube logo
关于 JoVE
概览领导团队博客JoVE 帮助中心
作者
出版流程编辑委员会范围与政策同行评审常见问题投稿
图书馆员
用户评价订阅访问资源图书馆顾问委员会常见问题
研究
JoVE JournalMethods CollectionsJoVE Encyclopedia of Experiments存档
教育
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab Manual教师资源中心教师网站
使用条款与条件
隐私政策
政策

相关概念视频

您也可能阅读

相关文章

通过共同作者、期刊和引用图与本文相关的文章。

排序
Same author

Facile construction of noble-metal-free Mn<sub>0.3</sub>Cd<sub>0.7</sub>S heterophase junction enriched with sulfur vacancies for efficient seawater splitting.

Journal of colloid and interface science·2026
Same author

Ultra-sensitive urine DNA methylation test enables early and accurate detection of bladder cancer.

Epigenomics·2026
Same author

Feasibility and preliminary safety of urologic telesurgery using Chinese surgical systems: a single-center exploratory case series.

BMC urology·2026
Same author

Optimization of bottom structure layout and key parameters in block caving: a case study of the Cukaru Peki Copper-Gold Mine in Serbia.

Scientific reports·2026
Same author

Integrative Mendelian Randomization and Pathomics Analysis Using Expression Quantitative Trait Loci and Genome-Wide Association Study Data Identifies Mismatch Repair Genes as Prognostic Biomarkers in Gastric Adenocarcinoma.

International journal of genomics·2026
Same author

Cardiac Cycle-Synchronous Intermittent Pneumatic Compression Therapy for Enhancing Peripheral Blood Perfusion.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2025

相关实验视频

Updated: Jan 10, 2026

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

5.2K

面向用户可通用的可穿戴式基于传感器的人类活动识别:多任务对比学习方法.

Pengyu Guo1, Masaya Nakayama2

  • 1Department of Electronic Engineering and Information Systems, The University of Tokyo, Tokyo 113-8654, Japan.

Sensors (Basel, Switzerland)
|November 27, 2025
PubMed
概括

本研究引入了一种新的多任务对比学习框架,以改善不同用户之间的人类活动识别 (HAR). 该方法提高了可穿戴传感器数据的概括性,使HAR系统更具可扩展性和适应性.

关键词:
人类活动识别 (HAR)相反的学习学习学习.多任务学习是多任务学习.监督的对比学习学习.用户通用的一般化.可穿戴式传感器传感器

更多相关视频

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment
06:49

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment

Published on: December 11, 2015

9.3K
Design and Analysis for Fall Detection System Simplification
08:05

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

11.1K

相关实验视频

Last Updated: Jan 10, 2026

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

5.2K
Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment
06:49

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment

Published on: December 11, 2015

9.3K
Design and Analysis for Fall Detection System Simplification
08:05

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

11.1K

科学领域:

  • 计算机科学 计算机科学
  • 机器学习 机器学习
  • 可穿戴技术可穿戴技术

背景情况:

  • 使用可穿戴传感器的人类活动识别 (HAR) 对个性化健康和无处不在的计算至关重要.
  • 当前的深度学习 HAR 模型在用户层面的概括上扎,阻碍了现实世界的应用.
  • 存在对HAR模型的需求,这些模型可以在不同的,看不见的用户中可靠地执行.

研究的目的:

  • 开发一种新的多任务对比学习框架,以提高HAR的用户级概括性.
  • 提高HAR系统的稳定性和可扩展性,以便在现实世界中部署.
  • 调查联合活动分类和用户意识的对比学习对HAR性能的影响.

主要方法:

  • 提出了一个多任务对比学习框架,将活动分类和监督对比目标结合起来.
  • 利用活动和用户标签来创建语义上丰富的对比对.
  • 采用了用户不可知推断策略来测试看不见的用户.

主要成果:

  • 在使用交叉用户评估的三个公共 HAR 数据集上,取得了与监督和自我监督基线可比的结果.
  • 通过废除研究证明了多任务培训和用户意识的对比监督的有效性.
  • 展示了改进的表示学习,以便在用户之间更好地泛化.

结论:

  • 拟议的多任务对比学习框架有效地提高了HAR的用户级概括性.
  • 这种方法为开发更具可扩展性和适应性的HAR系统提供了一个有希望的方向.
  • 用户意识到的对比监督是提高HAR模型对未见到用户的性能的一个关键组成部分.