Hybrid Consistency Training with Prototype Adaptation for Few-Shot Learning

TL;DR

This paper introduces a hybrid consistency training approach with prototype adaptation to enhance few-shot learning, addressing distribution shifts and limited labeled data, resulting in significant performance improvements across multiple datasets.

Contribution

It proposes a novel hybrid consistency training method combined with iterative prototype adaptation for better generalization in few-shot learning.

Findings

Achieves 2-5% improvement over state-of-the-art on five datasets.

Attains 7-8% higher accuracy in cross-domain FSL.

Demonstrates robustness against distribution shifts and limited labeled data.

Abstract

Few-Shot Learning (FSL) aims to improve a model's generalization capability in low data regimes. Recent FSL works have made steady progress via metric learning, meta learning, representation learning, etc. However, FSL remains challenging due to the following longstanding difficulties. 1) The seen and unseen classes are disjoint, resulting in a distribution shift between training and testing. 2) During testing, labeled data of previously unseen classes is sparse, making it difficult to reliably extrapolate from labeled support examples to unlabeled query examples. To tackle the first challenge, we introduce Hybrid Consistency Training to jointly leverage interpolation consistency, including interpolating hidden features, that imposes linear behavior locally and data augmentation consistency that learns robust embeddings against sample variations. As for the second challenge, we use unlabeled examples to iteratively normalize features and adapt prototypes, as opposed to commonly used one-time update, for more reliable prototype-based transductive inference. We show that our method generates a 2% to 5% improvement over the state-of-the-art methods with similar backbones on five FSL datasets and, more notably, a 7% to 8% improvement for more challenging cross-domain FSL.