BriefGPT.xyz
Mar, 2023
基于扩散的目标采样器用于无监督领域自适应
Diffusion-based Target Sampler for Unsupervised Domain Adaptation
HTML
PDF
Yulong Zhang, Shuhao Chen, Yu Zhang, Jiangang Lu
TL;DR
本文提出了一种基于扩散的目标采样器(DTS),使用分类条件信息生成高保真度与多样性伪目标样本,用于改进现有无监督领域自适应方法的传输性能。大量实验表明,该方法可以显著提高现有的无监督领域自适应方法的性能。
Abstract
Limited transferability hinders the performance of
deep learning
models when applied to new application scenarios. Recently,
unsupervised domain adaptation
(UDA) has achieved significant progress in addressing th
→