arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2203.11528stat.MLcs.LG
收藏

基于因果不变变换的分布外推广

Out-of-distribution Generalization with Causal Invariant Transformations

University of Chinese Academy of Sciences(中国科学院大学);Academy of Mathematics and Systems Science, Chinese Academy of Sciences(中国科学院数学与系统科学研究院);Huawei Noah’s Ark Lab(华为诺亚方舟实验室)

Ruoyu Wang, Mingyang Yi, Zhitang Chen, Shengyu Zhu

摘要:

在现实世界的应用程序中,学习在分发外(OOD)数据上表现良好的模型非常重要且可取。最近,因果关系已成为一个强大的工具,以解决面向对象的泛化问题,其思想依赖于因果机制,是跨领域的利益不变。为了利用通常未知的因果机制,现有的作品假设因果特征的线性形式,或者需要足够多和不同的训练域,这在实践中通常是限制性的。在这项工作中,我们将这些假设和解决OOD问题,而不显着恢复的因果功能。我们的方法基于修改非因果特征但保持因果部分不变的转换,这可以从先验知识中获得,也可以从多域场景中的训练数据中学习。在不变因果机制的设置下,我们从理论上证明了,如果所有这样的转换是可用的,那么我们可以学习一个极小极大最优模型跨域仅使用单域数据。注意到,知道这些因果不变变换的完整集合可能是不切实际的,我们进一步表明,它足以知道这些变换的一个子集。基于理论研究结果,提出了一种正则化的训练过程,以提高OOD的泛化能力。在人工和真实数据集上的大量实验结果验证了该算法的有效性,即使只有几个因果不变变换。

英文摘要:

In real-world applications, it is important and desirable to learn a model that performs well on out-of-distribution (OOD) data. Recently, causality has become a powerful tool to tackle the OOD generalization problem, with the idea resting on the causal mechanism that is invariant across domains of interest. To leverage the generally unknown causal mechanism, existing works assume a linear form of causal feature or require sufficiently many and diverse training domains, which are usually restrictive in practice. In this work, we obviate these assumptions and tackle the OOD problem without explicitly recovering the causal feature. Our approach is based on transformations that modify the non-causal feature but leave the causal part unchanged, which can be either obtained from prior knowledge or learned from the training data in the multi-domain scenario. Under the setting of invariant causal mechanism, we theoretically show that if all such transformations are available, then we can learn a minimax optimal model across the domains using only single domain data. Noticing that knowing a complete set of these causal invariant transformations may be impractical, we further show that it suffices to know only a subset of these transformations. Based on the theoretical findings, a regularized training procedure is proposed to improve the OOD generalization capability. Extensive experimental results on both synthetic and real datasets verify the effectiveness of the proposed algorithm, even with only a few causal invariant transformations.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑