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

Huang Qi Decoction Prevents BDL-Induced Liver Fibrosis Through Inhibition of Notch Signaling Activation.

The American journal of Chinese medicine·2017
Same author

Astragaloside IV Attenuates Podocyte Apoptosis Mediated by Endoplasmic Reticulum Stress through Upregulating Sarco/Endoplasmic Reticulum Ca<sup>2+</sup>-ATPase 2 Expression in Diabetic Nephropathy.

Frontiers in pharmacology·2017
Same author

A bio-chemical application of N-GQDs and g-C<sub>3</sub>N<sub>4</sub> QDs sensitized TiO<sub>2</sub> nanopillars for the quantitative detection of pcDNA3-HBV.

Biosensors & bioelectronics·2017
Same author

Clinical and imaging analysis of subclinical hemophilia combined with coxarthrosis: case report and literature review.

SpringerPlus·2016
Same author

On the summertime air quality and related photochemical processes in the megacity Shanghai, China.

The Science of the total environment·2016
Same author

Cuticular Wax Accumulation Is Associated with Drought Tolerance in Wheat Near-Isogenic Lines.

Frontiers in plant science·2016

相关实验视频

Updated: Jul 7, 2025

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

589

通过相互信息估计和最小化培训多源域适应网络.

Lisheng Wen1, Sentao Chen1, Mengying Xie2

  • 1Department of Computer Science, Shantou University, China.

Neural networks : the official journal of the International Neural Network Society
|December 21, 2023
PubMed
概括

本研究引入了一种新的多源域适应 (MSDA) 方法,通过最小化特征空间之间的相互信息. 这种方法有效地调整数据分布,以改善目标数据集的分类.

关键词:
凸起式优化的优化多源域调整多源域调整互助信息互助信息互助信息互助信息统计学学习 统计学学习

更多相关视频

Author Spotlight: Impact of Intergenic Interactions on Disease-Identifying Dark Biomarkers
03:37

Author Spotlight: Impact of Intergenic Interactions on Disease-Identifying Dark Biomarkers

Published on: March 1, 2024

761
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

405

相关实验视频

Last Updated: Jul 7, 2025

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

589
Author Spotlight: Impact of Intergenic Interactions on Disease-Identifying Dark Biomarkers
03:37

Author Spotlight: Impact of Intergenic Interactions on Disease-Identifying Dark Biomarkers

Published on: March 1, 2024

761
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

405

科学领域:

  • 机器学习 机器学习
  • 人工智能的人工智能
  • 计算机视觉 计算机视觉

背景情况:

  • 多源域调整 (MSDA) 在多个标记数据集上训练模型,用于未标记的目标数据.
  • 一个关键的挑战是适应跨数据集的不同联合分布.
  • 现有的方法在有效的分布对齐方面扎.

研究的目的:

  • 为多源域适应 (MSDA) 提出一种新的方法.
  • 为了在潜伏特征空间中对准源和目标联合分布.
  • 提高神经网络在未标记的目标数据上的性能.

主要方法:

  • 估计和最小化网络潜伏特征空间内的相互信息.
  • 制定相互信息估计作为一个凸的优化问题,以保证全球最佳.
  • 开发一种用于域调整的新型算法.

主要成果:

  • 拟议的算法在公共数据集上在统计上优于现有的方法.
  • 证明了源和目标联合分布的有效对齐.
  • 在未标记的目标数据上实现了优越的分类性能.

结论:

  • 尽量减少相互信息是MSDA的一个有效策略.
  • 拟议的凸式优化方法提供了一个全球最佳解决方案.
  • 该方法为神经网络的跨领域学习提供了显著的进步.