Semantics-driven Attentive Few-shot Learning over Clean and Noisy Samples

TL;DR

This paper introduces a semantic attention-based meta-learning approach for few-shot learning that effectively handles both clean and noisy samples by leveraging prior semantic knowledge to improve classification accuracy.

Contribution

It proposes semantically-conditioned feature and sample attention mechanisms to enhance meta-learner performance in few-shot learning, especially under noisy data conditions.

Findings

Semantic FSL model outperforms baselines on benchmark datasets.

The approach is robust to noisy training samples.

Semantic attention improves class discrimination in few-shot scenarios.

Abstract

Over the last couple of years few-shot learning (FSL) has attracted great attention towards minimizing the dependency on labeled training examples. An inherent difficulty in FSL is the handling of ambiguities resulting from having too few training samples per class. To tackle this fundamental challenge in FSL, we aim to train meta-learner models that can leverage prior semantic knowledge about novel classes to guide the classifier synthesis process. In particular, we propose semantically-conditioned feature attention and sample attention mechanisms that estimate the importance of representation dimensions and training instances. We also study the problem of sample noise in FSL, towards the utilization of meta-learners in more realistic and imperfect settings. Our experimental results demonstrate the effectiveness of the proposed semantic FSL model with and without sample noise.