Unsupervised domain adaptation (UDA) aims to learn a model for unlabeled data on a target domain by transferring knowledge from a labeled source domain. In the traditional UDA setting, labeled source data are assumed to be available for the use of model adaptation. Due to the increasing concerns for data privacy, source-free UDA is highly appreciated as a new UDA setting, where only a trained source model is assumed to be available, while the labeled source data remain private. However, exposing details of the trained source model for UDA use is prone to easily committed white-box attacks, which brings severe risks to the source tasks themselves. To address this issue, we advocate studying a subtly different setting, named Black-Box Unsupervised Domain Adaptation (B2UDA), where only the input-output interface of the source model is accessible in UDA; in other words, the source model itself is kept as a black-box one. To tackle the B2UDA task, we propose a simple yet effective method, termed Iterative Noisy Label Learning (IterNLL). IterNLL starts with getting noisy labels of the unlabeled target data from the black-box source model. It then alternates between learning improved target models from the target subset with more reliable labels and updating the noisy target labels. Experiments on benchmark datasets confirm the efficacy of our proposed method. Notably, IterNLL performs comparably with methods of the traditional UDA setting where the labeled source data are fully available.
翻译:不受监管域适应 (UDA) 的目的是通过从标签源域传输知识,学习目标域上未标签数据的模式。 在传统的 UDA 设置中, 标签源数据被假定为可用于模型适应。 由于对数据隐私的日益关注, 无源 UDA 被高度赞赏为一个新的 UDA 设置, 假设只有经过培训的源模型, 而标签源数据仍然保密 。 然而, 暴露UDA 使用经过培训的来源模式的细节很容易容易发生白箱攻击, 给源任务本身带来严重风险 。 为了解决这个问题, 我们主张研究一个亚异的设置, 名为 Black- Box Unurvedvised Domain 适应 (B2UDA), 因为在 UDA 中只能访问源模型的输入输出界面; 换句话说, 源模型本身被保留为黑箱 。 为了应对 B2UDA 任务, 我们提议了一个简单有效的方法, 称为 Iterative Label 学习 (Iber NLLLLL) 。 为了解决这个问题, 我们主张研究一个亚性目标源, 它开始从现在正在从透明标签 的标签 目标数据库中获取更可靠的目标源更新。