TapNet: Neural Network Augmented with Task-Adaptive Projection for Few-Shot Learning

TL;DR

TapNet introduces a neural network framework with task-adaptive projections and meta-learning, significantly improving few-shot learning performance across multiple datasets by dynamically conditioning features per task.

Contribution

The paper presents TapNet, a novel neural network architecture that incorporates task-specific feature projections and meta-learning for enhanced few-shot classification.

Findings

Achieves state-of-the-art accuracy on Omniglot, miniImageNet, and tieredImageNet.

Demonstrates effective generalization across diverse few-shot scenarios.

Utilizes a task-adaptive projection mechanism for rapid adaptation to new tasks.

Abstract

Handling previously unseen tasks after given only a few training examples continues to be a tough challenge in machine learning. We propose TapNets, neural networks augmented with task-adaptive projection for improved few-shot learning. Here, employing a meta-learning strategy with episode-based training, a network and a set of per-class reference vectors are learned across widely varying tasks. At the same time, for every episode, features in the embedding space are linearly projected into a new space as a form of quick task-specific conditioning. The training loss is obtained based on a distance metric between the query and the reference vectors in the projection space. Excellent generalization results in this way. When tested on the Omniglot, miniImageNet and tieredImageNet datasets, we obtain state of the art classification accuracies under various few-shot scenarios.