Instance-based Max-margin for Practical Few-shot Recognition

TL;DR

This paper introduces a practical few-shot learning setting leveraging unsupervised pretraining and proposes IbM2, an instance-based max-margin method that improves recognition performance in both traditional and new FSL scenarios.

Contribution

The paper presents a new practical FSL setting and a novel IbM2 method that enhances recognition accuracy across various tasks and pretraining methods.

Findings

IbM2 consistently improves performance over baselines.

The method works well in both traditional and practical FSL scenarios.

Significant gains observed across diverse datasets.

Abstract

In order to mimic the human few-shot learning (FSL) ability better and to make FSL closer to real-world applications, this paper proposes a practical FSL (pFSL) setting. pFSL is based on unsupervised pretrained models (analogous to human prior knowledge) and recognizes many novel classes simultaneously. Compared to traditional FSL, pFSL is simpler in its formulation, easier to evaluate, more challenging and more practical. To cope with the rarity of training examples, this paper proposes IbM2, an instance-based max-margin method not only for the new pFSL setting, but also works well in traditional FSL scenarios. Based on the Gaussian Annulus Theorem, IbM2 converts random noise applied to the instances into a mechanism to achieve maximum margin in the many-way pFSL (or traditional FSL) recognition task. Experiments with various self-supervised pretraining methods and diverse many- or few-way FSL tasks show that IbM2 almost always leads to improvements compared to its respective baseline methods, and in most cases the improvements are significant. With both the new pFSL setting and novel IbM2 method, this paper shows that practical few-shot learning is both viable and promising.