Finding Task-Relevant Features for Few-Shot Learning by Category Traversal

TL;DR

This paper introduces a Category Traversal Module that enhances metric-learning based few-shot learners by identifying task-relevant features across the entire support set, leading to significant performance improvements on standard benchmarks.

Contribution

The paper proposes a novel plug-and-play module for few-shot learning that considers the entire support set to identify relevant features, addressing limitations of previous independent class approaches.

Findings

Achieves 5-10% relative performance improvement on mini-ImageNet and tieredImageNet.

Competitive with recent state-of-the-art few-shot learning systems.

Demonstrates the effectiveness of cross-class feature traversal in few-shot tasks.

Abstract

Few-shot learning is an important area of research. Conceptually, humans are readily able to understand new concepts given just a few examples, while in more pragmatic terms, limited-example training situations are common in practice. Recent effective approaches to few-shot learning employ a metric-learning framework to learn a feature similarity comparison between a query (test) example, and the few support (training) examples. However, these approaches treat each support class independently from one another, never looking at the entire task as a whole. Because of this, they are constrained to use a single set of features for all possible test-time tasks, which hinders the ability to distinguish the most relevant dimensions for the task at hand. In this work, we introduce a Category Traversal Module that can be inserted as a plug-and-play module into most metric-learning based few-shot learners. This component traverses across the entire support set at once, identifying task-relevant features based on both intra-class commonality and inter-class uniqueness in the feature space. Incorporating our module improves performance considerably (5%-10% relative) over baseline systems on both mini-ImageNet and tieredImageNet benchmarks, with overall performance competitive with recent state-of-the-art systems.