Relational Embedding for Few-Shot Classification

TL;DR

This paper introduces a relational embedding approach for few-shot classification that uses self- and cross-correlational modules to learn structural and co-attentional patterns, improving performance on standard benchmarks.

Contribution

It presents a novel relational embedding network combining self- and cross-correlational modules for few-shot learning, demonstrating superior results over existing methods.

Findings

Achieves consistent improvements on miniImageNet, tieredImageNet, CUB-200-2011, and CIFAR-FS.

Introduces self-correlational and cross-correlational modules for relational pattern learning.

End-to-end training of the relational embedding network.

Abstract

We propose to address the problem of few-shot classification by meta-learning "what to observe" and "where to attend" in a relational perspective. Our method leverages relational patterns within and between images via self-correlational representation (SCR) and cross-correlational attention (CCA). Within each image, the SCR module transforms a base feature map into a self-correlation tensor and learns to extract structural patterns from the tensor. Between the images, the CCA module computes cross-correlation between two image representations and learns to produce co-attention between them. Our Relational Embedding Network (RENet) combines the two relational modules to learn relational embedding in an end-to-end manner. In experimental evaluation, it achieves consistent improvements over state-of-the-art methods on four widely used few-shot classification benchmarks of miniImageNet, tieredImageNet, CUB-200-2011, and CIFAR-FS.