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

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.

CVPR 2019

References in corpus (10)

• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• Prototypical Networks for Few-shot Learning
• A Simple Neural Attentive Meta-Learner
• Meta-SGD: Learning to Learn Quickly for Few-Shot Learning
• On First-Order Meta-Learning Algorithms
• Few-Shot Learning with Graph Neural Networks
• TADAM: Task dependent adaptive metric for improved few-shot learning
• Incremental Few-Shot Learning with Attention Attractor Networks
• Image Deformation Meta-Networks for One-Shot Learning
• AgileNet: Lightweight Dictionary-based Few-shot Learning

Cited by in corpus (21)

• Rethinking Few-Shot Image Classification: a Good Embedding Is All You Need?
• Rectifying the Shortcut Learning of Background for Few-Shot Learning
• DeepEMD: Differentiable Earth Mover's Distance for Few-Shot Learning
• MHFC: Multi-Head Feature Collaboration for Few-Shot Learning
• Few-Shot Object Detection with Attention-RPN and Multi-Relation Detector
• Meta-DETR: Image-Level Few-Shot Object Detection with Inter-Class Correlation Exploitation
• Relational Generalized Few-Shot Learning
• Feature Intertwiner for Object Detection
• Few-Shot Action Recognition with Compromised Metric via Optimal Transport
• Moving Indoor: Unsupervised Video Depth Learning in Challenging Environments
• Instance Credibility Inference for Few-Shot Learning
• Contrastive Prototype Learning with Augmented Embeddings for Few-Shot Learning
• Graph-based Interpolation of Feature Vectors for Accurate Few-Shot Classification
• Learning from the Past: Continual Meta-Learning via Bayesian Graph Modeling
• Prototype Completion with Primitive Knowledge for Few-Shot Learning
• ECKPN: Explicit Class Knowledge Propagation Network for Transductive Few-shot Learning
• Few-shot Open-set Recognition by Transformation Consistency
• Few-Shot Open-Set Recognition using Meta-Learning
• Extended Few-Shot Learning: Exploiting Existing Resources for Novel Tasks
• Reducing Overlearning through Disentangled Representations by Suppressing Unknown Tasks
• Finding Significant Features for Few-Shot Learning using Dimensionality Reduction