研究分类器随领域变化的泛化问题，突破传统静态模型局限。
Domain Generalization Under Posterior Drift
- 基于后验漂移假设构建决策理论框架
- 在语言与视觉任务中验证跨领域性能差异
- 适合关注动态适应性的研究人员
领域泛化（DG）旨在从多个有标签训练域推广到无标签测试域。现有主流基准数据集通常存在一个在所有域表现良好的单一分类器。本文研究一种根本不同的情形：各域满足后验漂移假设，即最优分类器可能随域显著变化。我们建立了一个面向后验漂移的领域泛化决策理论框架，并通过语言与视觉任务的实验探讨其实际意义。
原文摘要 · Abstract (English)
Domain generalization (DG) is the problem of generalizing from several distributions (or domains), for which labeled training data are available, to a new test domain for which no labeled data is available. For the prevailing benchmark datasets in DG, there exists a single classifier that performs well across all domains. In this work, we study a fundamentally different regime where the domains satisfy a \emph{posterior drift} assumption, in which the optimal classifier might vary substantially with domain. We establish a decision-theoretic framework for DG under posterior drift, and investigate the practical implications of this framework through experiments on language and vision tasks.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。