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

Selected Data About Geographic Locations01:25

Selected Data About Geographic Locations

23
Geographic Information Systems (GIS) rely on two core types of data: spatial data and attribute data.Spatial DataSpatial data defines the physical location of features within a coordinate system, typically expressed in terms of latitude and longitude. It provides precise positioning for elements like roads, rivers, or buildings.Attribute DataAttribute data complements spatial data by adding descriptive information about these features. For example, a road's spatial data includes its start and...
23

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

Identification of small-molecule HSF1 amplifiers by high content screening in protection of cells from stress induced injury.

Biochemical and biophysical research communications·2009
Same author

Nanowire transformation by size-dependent cation exchange reactions.

Nano letters·2009
Same author

Effect of haishengsu as an adjunct therapy for patients with advanced renal cell cancer: a randomized and placebo-controlled clinical trial.

Journal of alternative and complementary medicine (New York, N.Y.)·2009
Same author

Identification of inhibitors of HSF1 functional activity by high-content target-based screening.

Journal of biomolecular screening·2009
Same author

Antitumor effects of targeting hTERT lentivirus-mediated RNA interference against KB cell lines.

Oncology research·2009
Same author

Characteristics of emissive spectrum and the removal of nitric oxide in N2/02/NO plasma with argon additive.

Journal of environmental sciences (China)·2009

相关实验视频

Updated: May 28, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

455

H:通过空间约束抽样,可扩展网络和混合任务进行长尾分类.

Wenyi Zhao1, Wei Li1, Yongqin Tian2

  • 1School of Artificial Intelligence, Beijing University of Posts and Telecommunications, Beijing, 100876, China.

Neural networks : the official journal of the International Neural Network Society
|February 12, 2025
PubMed
概括

这项研究引入了一种新的长尾分类方法,通过空间约束抽样和混合任务增强特征提取. 该方法在基准数据集上取得了最先进的结果.

关键词:
平衡的特征是平衡的特征.混合任务是混合任务.长尾动物的分类 长尾动物的分类可扩展的网络可扩展性.空间限制采样采样

更多相关视频

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
08:25

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Published on: May 7, 2019

8.9K
A System for Tracking the Dynamics of Social Preference Behavior in Small Rodents
08:38

A System for Tracking the Dynamics of Social Preference Behavior in Small Rodents

Published on: November 21, 2019

7.5K

相关实验视频

Last Updated: May 28, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

455
Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
08:25

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Published on: May 7, 2019

8.9K
A System for Tracking the Dynamics of Social Preference Behavior in Small Rodents
08:38

A System for Tracking the Dynamics of Social Preference Behavior in Small Rodents

Published on: November 21, 2019

7.5K

科学领域:

  • 计算机视觉 计算机视觉
  • 机器学习 机器学习

背景情况:

  • 由于数据分布不平衡,长尾分类在创建明确的决策边界方面存在挑战.
  • 现有的方法很难有效地整合语义和纹理信息,以进行强大的特征提取.

研究的目的:

  • 为长尾分类开发一个端到端可训练的方法,以改善特征表示.
  • 通过整合语义一致性和纹理特征来增强模型捕获平衡特征的能力.

主要方法:

  • 提出了空间约束抽样策略,以提供具有代表性特征的模型.
  • 引入了一个可扩展的网络架构,用于动态功能调整.
  • 开发了一个混合任务,集成单模型分类和跨模型对比学习,以全面捕获特征.

主要成果:

  • 在CIFAR10-LT,CIFAR100-LT,ImageNet-LT和iNaturalist 2018数据集上实现了最先进的性能.
  • 证明了高层次语义和低层次纹理特征的有效集成.
  • 启用端到端培训,克服多阶段优化约束.

结论:

  • 拟议的方法为长尾分类提供了一种新且有效的解决方案.
  • 空间约束抽样,可扩展网络和混合任务的结合导致了优越的特征学习.
  • 这种方法为更强大,更有效的长尾分类模型铺平了道路.