Few-Shot Learning with Metric-Agnostic Conditional Embeddings

TL;DR

This paper introduces a flexible, class-conditioned embedding architecture for few-shot learning that outperforms existing methods by learning what features are important for comparison, achieving state-of-the-art results.

Contribution

It proposes a novel architecture that conditions class representations on each target image and learns comparison strategies rather than using fixed metrics.

Findings

Achieves state-of-the-art performance on Caltech-UCSD birds dataset.

Outperforms traditional metric-learning approaches.

Demonstrates the effectiveness of class-conditioned 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.