Gaussian Prototypical Networks for Few-Shot Learning on Omniglot

TL;DR

This paper introduces Gaussian prototypical networks, an extension of prototypical networks that models uncertainty with Gaussian covariances, leading to improved few-shot classification performance on Omniglot.

Contribution

The paper proposes Gaussian prototypical networks that incorporate uncertainty estimates into the embedding space, enhancing classification accuracy over standard prototypical networks.

Findings

Achieved state-of-the-art results in 1-shot and 5-shot classification on Omniglot.

Gaussian prototypical networks outperform vanilla prototypical networks with similar parameters.

Down-sampling training images further improves performance, indicating robustness in noisier datasets.

Abstract

We propose a novel architecture for $k$-shot classification on the Omniglot dataset. Building on prototypical networks, we extend their architecture to what we call Gaussian prototypical networks. Prototypical networks learn a map between images and embedding vectors, and use their clustering for classification. In our model, a part of the encoder output is interpreted as a confidence region estimate about the embedding point, and expressed as a Gaussian covariance matrix. Our network then constructs a direction and class dependent distance metric on the embedding space, using uncertainties of individual data points as weights. We show that Gaussian prototypical networks are a preferred architecture over vanilla prototypical networks with an equivalent number of parameters. We report state-of-the-art performance in 1-shot and 5-shot classification both in 5-way and 20-way regime (for 5-shot 5-way, we are comparable to previous state-of-the-art) on the Omniglot dataset. We explore artificially down-sampling a fraction of images in the training set, which improves our performance even further. We therefore hypothesize that Gaussian prototypical networks might perform better in less homogeneous, noisier datasets, which are commonplace in real world applications.