Active Few-Shot Classification: a New Paradigm for Data-Scarce Learning Settings

TL;DR

This paper introduces Active Few-Shot Classification (AFSC), a new approach for classifying small unlabeled datasets with limited labels, showing significant accuracy improvements over existing methods in data-scarce scenarios.

Contribution

The paper proposes a novel AFSC framework combining statistical inference and active learning, and demonstrates its effectiveness on standard vision benchmarks.

Findings

Up to 10% accuracy gain over state-of-the-art TFSC methods

Effective in data-scarce learning settings

Introduces a new paradigm for label-efficient classification

Abstract

We consider a novel formulation of the problem of Active Few-Shot Classification (AFSC) where the objective is to classify a small, initially unlabeled, dataset given a very restrained labeling budget. This problem can be seen as a rival paradigm to classical Transductive Few-Shot Classification (TFSC), as both these approaches are applicable in similar conditions. We first propose a methodology that combines statistical inference, and an original two-tier active learning strategy that fits well into this framework. We then adapt several standard vision benchmarks from the field of TFSC. Our experiments show the potential benefits of AFSC can be substantial, with gains in average weighted accuracy of up to 10% compared to state-of-the-art TFSC methods for the same labeling budget. We believe this new paradigm could lead to new developments and standards in data-scarce learning settings.