论文标题
目标条件表示独立性（TCRI）；从域不变到域代表
Target Conditioned Representation Independence (TCRI); From Domain-Invariant to Domain-General Representations
论文作者
论文摘要
我们提出了目标条件表示独立性（TCRI）目标，以实现域的概括。 TCRI解决了由于不完全限制而导致的现有域泛化方法的局限性。具体而言，TCRI实现了由有条件独立约束的正规化器，这些限制足以严格学习不变机制的完整集合，我们表明，这对于域的概括是必要的，足以满足。从经验上讲，我们表明TCRI在合成和现实世界中均有效。 TCRI的平均准确性具有竞争力，同时以最差的域准确性优于它们，表明所需的跨域稳定性。
We propose a Target Conditioned Representation Independence (TCRI) objective for domain generalization. TCRI addresses the limitations of existing domain generalization methods due to incomplete constraints. Specifically, TCRI implements regularizers motivated by conditional independence constraints that are sufficient to strictly learn complete sets of invariant mechanisms, which we show are necessary and sufficient for domain generalization. Empirically, we show that TCRI is effective on both synthetic and real-world data. TCRI is competitive with baselines in average accuracy while outperforming them in worst-domain accuracy, indicating desired cross-domain stability.
