基于原型多层学习的半监督域自适应（英文版）

英文标题：Semi-supervised Domain Adaptation via Prototype-based Multi-level Learning中文摘要：本文提出了一种基于多级原型学习的半监督域自适应框架，采用伪标签聚合和交叉域对齐损失方法，以及通过原型相似度和线性分类器提升目标特征表示的判别性学习，实现了在三个数据集上卓越的 SSDA 性能。英文摘要：In semi-supervised domain adaptation (SSDA), a few labeled target samples ofeach class help the model to transfer knowledge representation from the fullylabeled source domain to the target domain. Many existing methods ignore thebenefits of making full use of the labeled target samples from multi-level. Tomake be