Recent domain generalization (DG) approaches typically use the hypothesis learned on source domains for inference on the unseen target domain. However, such a hypothesis can be arbitrarily far from the optimal one for the target domain, induced by a gap termed ``adaptivity gap''. Without exploiting the domain information from the unseen test samples, adaptivity gap estimation and minimization are intractable, which hinders us to robustify a model to any unknown distribution. In this paper, we first establish a generalization bound that explicitly considers the adaptivity gap. Our bound motivates two strategies to reduce the gap: the first one is ensembling multiple classifiers to enrich the hypothesis space, then we propose effective gap estimation methods for guiding the selection of a better hypothesis for the target. The other method is minimizing the gap directly by adapting model parameters using online target samples. We thus propose \textbf{Domain-specific Risk Minimization (DRM)}. During training, DRM models the distributions of different source domains separately; for inference, DRM performs online model steering using the source hypothesis for each arriving target sample. Extensive experiments demonstrate the effectiveness of the proposed DRM for domain generalization with the following advantages: 1) it significantly outperforms competitive baselines on different distributional shift settings; 2) it achieves either comparable or superior accuracies on all source domains compared to vanilla empirical risk minimization; 3) it remains simple and efficient during training, and 4) it is complementary to invariant learning approaches.
翻译:最近的域概略( DG) 方法通常使用在源域上获得的假设,用于对隐蔽目标域进行推断。然而,这种假设可能任意地与目标域的最佳假设相去甚远,因为“适应性差距”是一个空白。另一个方法是直接通过利用在线目标样本调整模型参数来缩小差距。因此,我们提议了\ textbf{Domain特定风险最小化(DRM)}。在培训期间,DRM将不同源域的分布分别地分开考虑。根据推断,DRM将利用每个目标样本的来源假设来进行在线模式指导,以丰富假设空间,然后我们提出有效的差距估计方法来指导选择更好的目标假设。另一个方法是直接通过使用在线目标样本调整模型参数来缩小差距。因此,我们提议了\ textb{Domain 特定风险最小化(DRM)}。在培训期间,DRMM将不同源域的分布分别作为模型;在每次到达目标样本样本样本样本时,DRM将使用源式指导。 广泛实验显示DRM 基准下的所有范围的升级,在一般范围中,在学习期间,在风险评估中,在不同的基准中,在不同的基准下,在风险评估中,在风险评估中,在风险评估中,在风险评估下,在风险评估下,在进行。