论文标题
Decor：不变学习和OOD概括的环境分区
Decorr: Environment Partitioning for Invariant Learning and OOD Generalization
论文作者
论文摘要
旨在识别多种环境的一致预测因子的不变学习方法在分布（OOD）概括方面已获得突出。但是，当数据不固有的环境时，从业人员必须手动定义它们。这种环境分配 - 将培训数据集分割为环境，对不变的学习功效产生了至关重要的影响，但仍未引起人们的注意。适当的环境分配可以扩大不变学习的适用性并提高其性能。在本文中，我们建议通过隔离低相关数据子集将数据集划分为多种环境。通过使用合成和真实数据的实验，我们的Decorr方法与不变学习结合使用了卓越的性能。 Decor减轻了虚假相关性的问题，有助于识别稳定的预测因子，并扩大了不变学习方法的适用性。
Invariant learning methods, aimed at identifying a consistent predictor across multiple environments, are gaining prominence in out-of-distribution (OOD) generalization. Yet, when environments aren't inherent in the data, practitioners must define them manually. This environment partitioning--algorithmically segmenting the training dataset into environments--crucially affects invariant learning's efficacy but remains underdiscussed. Proper environment partitioning could broaden the applicability of invariant learning and enhance its performance. In this paper, we suggest partitioning the dataset into several environments by isolating low-correlation data subsets. Through experiments with synthetic and real data, our Decorr method demonstrates superior performance in combination with invariant learning. Decorr mitigates the issue of spurious correlations, aids in identifying stable predictors, and broadens the applicability of invariant learning methods.
