发表机构
Toyota Technological Institute at Chicago; University of Waterloo（芝加哥丰田技术学院; 滑铁卢大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究分层域泛化问题，从有限观察区域外推到整个实例空间，以任意域层次结构取代独立同分布采样，揭示核心障碍在于训练/测试域划分，表明现代泛化理论应将域结构视为关键因素。
AI 中文摘要
我们将分层域泛化作为一个从有限观察区域外推到整个实例空间的问题进行研究，用任意域层次结构取代独立同分布采样。我们表明，核心障碍不仅在于假设类的复杂性，还在于揭示证据的训练/测试域划分。特别是，无论类有多小或训练规模有多大，某些划分都会导致对某些目标的泛化失败。这些结果表明，现代泛化理论必须将域结构视为一等对象。
英文摘要
We study hierarchical domain generalization as a problem of extrapolation from finite observed regions to an entire instance space, replacing i.i.d. sampling with arbitrary domain hierarchies. We show that the central obstruction is not only the complexity of the hypothesis class, but the train/test domain partition through which evidence is revealed. In particular, no matter how small the class or how large the training size, some partition makes generalization fail for some target. These results suggest that modern generalization theory must treat domain structure as a first-class object.
