Few-Shot Learning with Metric-Agnostic Conditional Embeddings

Abstract

Learning high quality class representations from few examples is a key problem in metric-learning approaches to few-shot learning. To accomplish this, we introduce a novel architecture where class representations are conditioned for each few-shot trial based on a target image. We also deviate from traditional metric-learning approaches by training a network to perform comparisons between classes rather than relying on a static metric comparison. This allows the network to decide what aspects of each class are important for the comparison at hand. We find that this flexible architecture works well in practice, achieving state-of-the-art performance on the Caltech-UCSD birds fine-grained classification task.

References in corpus (1)

• Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)

Cited by in corpus (34)

• A Closer Look at Few-shot Classification
• A Two-Stage Approach to Few-Shot Learning for Image Recognition
• Further advantages of data augmentation on convolutional neural networks
• Cross-Domain Few-Shot Classification via Learned Feature-Wise Transformation
• Bayesian Meta-Learning for the Few-Shot Setting via Deep Kernels
• BOIL: Towards Representation Change for Few-shot Learning
• Transductive Episodic-Wise Adaptive Metric for Few-Shot Learning
• Negative Margin Matters: Understanding Margin in Few-shot Classification
• Design of a Graphical User Interface for Few-Shot Machine Learning Classification of Electron Microscopy Data
• Self-Supervised Learning For Few-Shot Image Classification
• Adversarial Feature Hallucination Networks for Few-Shot Learning
• Weakly-supervised Compositional FeatureAggregation for Few-shot Recognition
• FS-BAN: Born-Again Networks for Domain Generalization Few-Shot Classification
• Instance Credibility Inference for Few-Shot Learning
• Bayesian Few-Shot Classification with One-vs-Each Pólya-Gamma Augmented Gaussian Processes
• Multi-Objective Meta Learning
• Generalized Adaptation for Few-Shot Learning
• Regularizing Meta-Learning via Gradient Dropout
• Large Margin Mechanism and Pseudo Query Set on Cross-Domain Few-Shot Learning
• Few-shot Learning for Domain-specific Fine-grained Image Classification
• Regression Networks for Meta-Learning Few-Shot Classification
• Shot in the Dark: Few-Shot Learning with No Base-Class Labels
• How to trust unlabeled data? Instance Credibility Inference for Few-Shot Learning
• One-Click Annotation with Guided Hierarchical Object Detection
• Few-shot learning via tensor hallucination
• Combining Domain-Specific Meta-Learners in the Parameter Space for Cross-Domain Few-Shot Classification
• GCCN: Global Context Convolutional Network
• Will Multi-modal Data Improves Few-shot Learning?
• When the Few Outweigh the Many: Illicit Content Recognition with Few-Shot Learning
• Multi-level Similarity Learning for Low-Shot Recognition
• Novelty-Prepared Few-Shot Classification
• Proxy Network for Few Shot Learning
• MetaMix: Improved Meta-Learning with Interpolation-based Consistency Regularization
• Reinforced Attention for Few-Shot Learning and Beyond