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

Natural Selection and Adaptation01:15

Natural Selection and Adaptation

177
Natural selection, a fundamental concept in evolutionary biology, is the mechanism by which evolution is driven, favoring organisms that are best adapted to their environments. This process enhances their chances of survival and reproduction. Adaptation, a key outcome of this process, involves genetic modifications that optimize an organism's functionality under specific environmental challenges, such as extreme cold or thinner air at high altitudes.
Beyond physical adaptations,...
177

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

A Predictive Framework for Acute Liver Failure Outcomes: Insights from Nonbiological Artificial Liver Treatment.

Clinical laboratory·2026
Same author

From polyamine metabolism genes to immunoproteasome subunits: a transcriptome-based screening of candidate biomarkers in Crohn's disease.

Scientific reports·2026
Same author

Factors influencing the intention to use "Internet+Nursing Service" among enterostomy patients based on an extended UTAUT2 model.

Scientific reports·2026
Same author

Lymphocyte subsets in untreated thalassemia patients: differences by genotype and age.

Frontiers in immunology·2026
Same author

Characteristic wavelength selection for rice blast based on hyperspectral remote sensing and deep convolutional neural networks.

Pest management science·2026
Same author

Serum proteomic profiling of sepsis patients reveals a protein-based diagnostic model, with metabolomic insights into carbapenem-resistant <i>Klebsiella pneumoniae</i> infection.

Frontiers in immunology·2026

相关实验视频

Updated: Jun 10, 2025

Visualizing Visual Adaptation
04:43

Visualizing Visual Adaptation

Published on: April 24, 2017

8.9K

多源完全测试时间适应多源完全测试时间适应

Yuntao Du1, Siqi Luo2, Yi Xin2

  • 1Beijing Institute for General Artificial Intelligence (BIGAI), Beijing, China.

Neural networks : the official journal of the International Neural Network Society
|October 11, 2024
PubMed
概括
此摘要是机器生成的。

本研究介绍了多源完全测试时间的适应,以改善深度学习模型的概括性. 这种新的方法有效地将多个模型适应新的数据分布,在多样化的测试样本上提高性能.

关键词:
域名适应领域适应测试时间适应.转移学习转移学习

更多相关视频

Computerized Adaptive Testing System of Functional Assessment of Stroke
05:21

Computerized Adaptive Testing System of Functional Assessment of Stroke

Published on: January 7, 2019

5.8K
Adaptation of a Haptic Robot in a 3T fMRI
08:16

Adaptation of a Haptic Robot in a 3T fMRI

Published on: October 4, 2011

9.7K

相关实验视频

Last Updated: Jun 10, 2025

Visualizing Visual Adaptation
04:43

Visualizing Visual Adaptation

Published on: April 24, 2017

8.9K
Computerized Adaptive Testing System of Functional Assessment of Stroke
05:21

Computerized Adaptive Testing System of Functional Assessment of Stroke

Published on: January 7, 2019

5.8K
Adaptation of a Haptic Robot in a 3T fMRI
08:16

Adaptation of a Haptic Robot in a 3T fMRI

Published on: October 4, 2011

9.7K

科学领域:

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

背景情况:

  • 深度神经网络 (DNN) 在许多领域都表现出色,但当测试数据分布与训练数据不同时,它们在概括方面遇到了困难.
  • 现有的完全测试时间适应方法使用单个训练模型,限制了适应的潜在信息.
  • 现实世界的场景通常涉及多个训练有素的模型,这些模型可以为适应提供互补的见解.

研究的目的:

  • 通过提出多源完全测试时间调整的问题来解决单一模型调整的局限性.
  • 开发一种利用多个预训练模型的方法,以便更好地适应未标记的测试数据.
  • 提高深度学习模型在域移动场景中的概括能力.

主要方法:

  • 引入了多源完全测试时间适应的新问题.
  • 开发了一种使用加权聚合方案来优先考虑相关模型的方法.
  • 结合了两个未经监督的损失,用于联合调整多个模型,使用在线未标记样本.

主要成果:

  • 与基线方法相比,拟议的方法在三个图像分类数据集上显示出更高的性能.
  • 权重聚合适应性地赋予更合适的模型更高的相关性.
  • 使用无监督损失的联合适应有效地利用了来自多个模型的互补信息.

结论:

  • 多源完全测试时间适应为改善模型概括提供了一个有希望的方向.
  • 建议的加权聚合和无监督损失方法有效地适应多个模型.
  • 这种方法在数据分布变化的场景中提高了稳定性和性能.