Given a causal graph representing the data-generating process shared across different domains/distributions, enforcing sufficient graph-implied conditional independencies can identify domain-general (non-spurious) feature representations. For the standard input-output predictive setting, we categorize the set of graphs considered in the literature into two distinct groups: (i) those in which the empirical risk minimizer across training domains gives domain-general representations and (ii) those where it does not. For the latter case (ii), we propose a novel framework with regularizations, which we demonstrate are sufficient for identifying domain-general feature representations without a priori knowledge (or proxies) of the spurious features. Empirically, our proposed method is effective for both (semi) synthetic and real-world data, outperforming other state-of-the-art methods in average and worst-domain transfer accuracy.

翻译：给定一个表示跨不同领域/分布共享数据生成过程的因果图，强制实施图隐含的条件独立性可以识别领域通用（非虚假）的特征表示。针对标准的输入输出预测设定，我们将文献中考虑的因果图集合分为两类：(i) 跨训练领域的经验风险最小化器能够生成领域通用表示的情形，以及(ii) 无法生成领域通用表示的情形。对于后一种情形(ii)，我们提出了一种带有正则化的新框架，并证明该框架足以在无需先验知识（或代理变量）了解虚假特征的情况下识别领域通用的特征表示。实验表明，我们的方法在（半）合成数据和真实数据上均有效，在平均和最差领域迁移准确率上均优于其他最先进方法。