Interventional Few-Shot Learning
It addresses performance limitations in few-shot learning for AI applications by introducing a causal approach that can enhance existing methods, though it is incremental in building on causal theory.
The paper identifies pre-trained knowledge as a confounder limiting few-shot learning performance and proposes Interventional Few-Shot Learning (IFSL), a novel paradigm based on causal intervention, which achieves new state-of-the-art results on benchmarks like miniImageNet and CUB with 1-/5-shot tasks.
We uncover an ever-overlooked deficiency in the prevailing Few-Shot Learning (FSL) methods: the pre-trained knowledge is indeed a confounder that limits the performance. This finding is rooted from our causal assumption: a Structural Causal Model (SCM) for the causalities among the pre-trained knowledge, sample features, and labels. Thanks to it, we propose a novel FSL paradigm: Interventional Few-Shot Learning (IFSL). Specifically, we develop three effective IFSL algorithmic implementations based on the backdoor adjustment, which is essentially a causal intervention towards the SCM of many-shot learning: the upper-bound of FSL in a causal view. It is worth noting that the contribution of IFSL is orthogonal to existing fine-tuning and meta-learning based FSL methods, hence IFSL can improve all of them, achieving a new 1-/5-shot state-of-the-art on \textit{mini}ImageNet, \textit{tiered}ImageNet, and cross-domain CUB. Code is released at https://github.com/yue-zhongqi/ifsl.