论文标题
基于损失的顺序学习，以适应主动域
Loss-based Sequential Learning for Active Domain Adaptation
论文作者
论文摘要
主动领域适应性（ADA）研究主要解决了查询选择，同时遵循现有的域适应策略。但是，我们认为，不仅要考虑查询选择标准，还要考虑为ADA场景设计的域适应策略至关重要。本文介绍了考虑域类型（源/目标）或标记度（标记/未标记）的顺序学习。我们首先仅在通过基于损失的查询选择获得的标记目标样本上训练模型。当在域移动下应用基于损耗的查询选择时，无用的高损失样本逐渐增加，并且标记的样本多样性变得较低。为了解决这些问题，我们通过利用损失预测来充分利用未标记目标域的伪标签。我们进一步鼓励伪标签具有低自我内向和多样化的班级分布。我们的模型在各种基准数据集中大大优于先前的方法以及基线模型。
Active domain adaptation (ADA) studies have mainly addressed query selection while following existing domain adaptation strategies. However, we argue that it is critical to consider not only query selection criteria but also domain adaptation strategies designed for ADA scenarios. This paper introduces sequential learning considering both domain type (source/target) or labelness (labeled/unlabeled). We first train our model only on labeled target samples obtained by loss-based query selection. When loss-based query selection is applied under domain shift, unuseful high-loss samples gradually increase, and the labeled-sample diversity becomes low. To solve these, we fully utilize pseudo labels of the unlabeled target domain by leveraging loss prediction. We further encourage pseudo labels to have low self-entropy and diverse class distributions. Our model significantly outperforms previous methods as well as baseline models in various benchmark datasets.
