.png){kind=icon}
返回
DFA: Decoupling Feature Alignment for Unsupervised Domain Adaptation
.png){kind=icon}
PRE
AI
DOI：10.1109/JIOT.2024.3423794.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
A prevailing assumption in existing deep learning research posits that data across source and target domains adhere to the independent and identically distributed (i.i.d.) assumption. However, this assumption often proves inadequate in real-world scenarios, leading to significant performance degradation when models encounter data with divergent distributions. To address this challenge, a novel unsupervised domain adaptation (UDA) algorithm, decoupling feature alignment (DFA), is introduced. The approach begins with the establishment of a robust theoretical framework, serving as the foundation for the mean-covariance adjustment feature alignment (MCAFA) algorithm. Simultaneously, a data decoupling (DD) module is introduced, effectively segregating target domain data into two subsets: one that mirrors the source domain and another that diverges markedly. Furthermore, a multidimensional alignment module is employed, leveraging the MCAFA algorithm and the DD module to align target data with source data across various layers and categories. Comprehensive evaluations on multiple data sets underscore the superiority of DFA.
Keyword:
Adaptation models
Training
Task analysis
Internet of Things
Data models
Neural networks
Deep learning
Data decoupling (DD)
feature alignment
independent and identically distributed (i.i.d.) assumption
multidimensional alignment (MDA)
unsupervised domain adaptation (UDA)
期刊
IF:
8.9
论文数:
1.4W
被引数:
7.8W
机构
引用论文
GAN-based data augmentation and pseudo-label refinement with holistic features for unsupervised domain adaptation person re-identification基于GAN的数据增强和具有整体特征的伪标签精化，用于无监督领域适应人员重新识别
PREAI
分享收藏
Shape-Consistent One-Shot Unsupervised Domain Adaptation for Rail Surface Defect Segmentation形状一致的一次无监督域自适应钢轨表面缺陷分割
PREAI
分享收藏Adversarial unsupervised domain adaptation for cross scenario waveform recognition.png)
SIGNAL PROCESSING
SIGNAL PROCESSINGIF3.6
PREAI
分享收藏Deadline-Aware Scheduling Policies for Bluetooth Networks in Industrial Communications工业通信中蓝牙网络的截止时间感知调度策略
PREAI
分享收藏ClassyNet: Class-Aware Early-Exit Neural Networks for Edge DevicesClassyNet: 用于边缘设备的类感知早期退出神经网络
PREAI
分享收藏
查看更多内容