学习可迁移的概念原型用于可解释的无监督域适应
Learning Transferable Conceptual Prototypes for Interpretable Unsupervised Domain Adaptation
AI总结：
本文提出可迁移概念原型学习（TCPL），通过层次化原型模块和融合原型信息的伪标签策略，实现无监督域适应的内在可解释性，同时提升知识迁移与决策性能。
摘要：
尽管深度神经网络在无监督域适应（UDA）方面取得了巨大进展，但当前的UDA模型是不透明的，无法提供可靠的解释，这限制了它们在需要安全且可控模型决策的场景中的应用。目前，大量工作集中于设计具有充分数据标注的深度可解释方法，而只有少数方法考虑了分布偏移问题。大多数现有的可解释UDA方法是事后（post-hoc）的，无法促进模型学习过程以提升性能。在本文中，我们提出了一种内在可解释的方法，名为可迁移概念原型学习（TCPL），它能够同时解释和改进UDA中的知识迁移与决策过程。为实现这一目标，我们设计了一个层次化原型模块，该模块将源域的类别基本概念迁移到目标域，并学习域共享的原型以解释底层推理过程。借助学习到的可迁移原型，我们设计了一种自预测一致的伪标签策略，该策略融合了置信度、预测和原型信息，用于选择合适的目标样本进行伪标注，并逐步缩小域差距。综合实验表明，所提出的方法不仅能提供有效且直观的解释，而且性能优于先前的最先进方法。
英文摘要：
Despite the great progress of unsupervised domain adaptation (UDA) with the deep neural networks, current UDA models are opaque and cannot provide promising explanations, limiting their applications in the scenarios that require safe and controllable model decisions. At present, a surge of work focuses on designing deep interpretable methods with adequate data annotations and only a few methods consider the distributional shift problem. Most existing interpretable UDA methods are post-hoc ones, which cannot facilitate the model learning process for performance enhancement. In this paper, we propose an inherently interpretable method, named Transferable Conceptual Prototype Learning (TCPL), which could simultaneously interpret and improve the processes of knowledge transfer and decision-making in UDA. To achieve this goal, we design a hierarchically prototypical module that transfers categorical basic concepts from the source domain to the target domain and learns domain-shared prototypes for explaining the underlying reasoning process. With the learned transferable prototypes, a self-predictive consistent pseudo-label strategy that fuses confidence, predictions, and prototype information, is designed for selecting suitable target samples for pseudo annotations and gradually narrowing down the domain gap. Comprehensive experiments show that the proposed method can not only provide effective and intuitive explanations but also outperform previous state-of-the-arts.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
