Few-shot learning with N-way K-shot scheme is an open challenge in machine learning. Many approaches have been proposed to tackle this problem, e.g., the Matching Networks and CLIP-Adapter. Despite that these approaches have shown significant progress, the mechanism of why these methods succeed has not been well explored. In this paper, we interpret these few-shot learning methods via causal mechanism. We show that the existing approaches can be viewed as specific forms of front-door adjustment, which is to remove the effects of confounders. Based on this, we introduce a general causal method for few-shot learning, which considers not only the relationship between examples but also the diversity of representations. Experimental results demonstrate the superiority of our proposed method in few-shot classification on various benchmark datasets. Code is available in the supplementary material.

翻译：N-way K-shot 方案少见的学习是机器学习的一个公开挑战。 已经提出了许多方法来解决这个问题, 例如匹配网络和CLIP-Adapter。 尽管这些方法已经取得了显著进展,但是这些方法取得成功的机制还没有很好地探索。 在本文中,我们通过因果机制来解释这些少见的学习方法。 我们表明,现有的方法可以被视为前门调整的具体形式,即消除混乱者的影响。 在此基础上,我们引入了一种一般的因果方法,用于少见的学习,不仅考虑实例之间的关系,而且考虑表达方式的多样性。 实验结果表明,我们所提议的方法在各种基准数据集的几发分类中具有优势。 守则可以在补充材料中找到。