Domain-Specific Bias Filtering for Single Labeled Domain Generalization
Domain-Specific Bias Filtering for Single Labeled Domain Generalization
复制标题
DOI：
10.1007/s11263-022-01712-7
复制
发表时间：
2021-10
影响因子：
19.5
通讯作者：
Junkun Yuan;Xu Ma;Defang Chen;Kun Kuang;Fei Wu;Lanfen Lin
中科院分区：
文献类型：
--
作者：
Junkun Yuan;Xu Ma;Defang Chen;Kun Kuang;Fei Wu;Lanfen Lin
传统领域泛化（CDG）利用多个标记的源数据集来训练未见过的目标领域的可泛化模型。然而，由于昂贵的标注成本，在实际应用中很难满足对所有源数据进行标注的要求。在本文中，我们研究了一种仅标记一个源域的单标记域泛化（SLDG）任务，该任务比 CDG 任务更实用且更具挑战性。 SLDG 任务的一个主要障碍是可辨别性泛化偏差：标记源数据集中的判别性信息可能包含特定领域的偏差，从而限制了训练模型的泛化。为了解决这一具有挑战性的任务，我们提出了一种称为特定领域偏差过滤（DSBF）的新颖框架，该框架使用标记的源数据初始化判别模型，然后使用未标记的源数据过滤掉其特定领域的偏差以提高泛化能力。我们将过滤过程分为（1）通过基于 k 均值聚类的语义特征重新提取进行特征提取器去偏，以及（2）通过注意力引导的语义特征投影进行分类器校正。 DSBF 统一了对标记和未标记源数据的探索，以增强训练模型的区分性和泛化性，从而产生高度泛化的模型。我们进一步提供理论分析来验证所提出的特定领域偏差过滤过程。对多个数据集的大量实验表明 DSBF 在处理具有挑战性的 SLDG 任务和 CDG 任务方面具有卓越的性能。
Conventional Domain Generalization (CDG) utilizes multiple labeled source datasets to train a generalizable model for unseen target domains. However, due to expensive annotation costs, the requirements of labeling all the source data are hard to be met in real-world applications. In this paper, we investigate a Single Labeled Domain Generalization (SLDG) task with only one source domain being labeled, which is more practical and challenging than the CDG task. A major obstacle in the SLDG task is the discriminability-generalization bias: the discriminative information in the labeled source dataset may contain domain-specific bias, constraining the generalization of the trained model. To tackle this challenging task, we propose a novel framework called Domain-Specific Bias Filtering (DSBF), which initializes a discriminative model with the labeled source data and then filters out its domain-specific bias with the unlabeled source data for generalization improvement. We divide the filtering process into (1) feature extractor debiasing via k-means clustering-based semantic feature re-extraction and (2) classifier rectification through attention-guided semantic feature projection. DSBF unifies the exploration of the labeled and the unlabeled source data to enhance the discriminability and generalization of the trained model, resulting in a highly generalizable model. We further provide theoretical analysis to verify the proposed domain-specific bias filtering process. Extensive experiments on multiple datasets show the superior performance of DSBF in tackling both the challenging SLDG task and the CDG task.