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. A common finding in the DG literature is that it is difficult to outperform empirical risk minimization (ERM) on the pooled training data. In this work, we argue that this finding has primarily been reported for datasets satisfying a \emph{covariate shift} assumption. When the dataset satisfies a \emph{posterior drift} assumption instead, we show that ``domain-informed ERM,'' wherein feature vectors are augmented with domain-specific information, outperforms pooling ERM. These claims are supported by a theoretical framework and experiments on language and vision tasks.

翻译：领域泛化（DG）旨在从多个已有标注训练数据的分布（或领域）中学习，并泛化到没有任何标注数据的新测试领域。领域泛化文献中一个普遍发现是：在合并训练数据上，很难超越经验风险最小化（ERM）的性能。本研究表明，这一结论主要基于满足**协变量偏移**假设的数据集。当数据集满足**后验漂移**假设时，我们证明"领域感知的ERM"——即用领域特定信息增强特征向量的方法——能够超越合并数据的ERM。这些结论通过理论框架以及在语言和视觉任务上的实验得到了验证。