Analyzing the Benefits of Prototypes for Semi-Supervised Category Learning

TL;DR

This paper investigates how prototype-based representations can enhance semi-supervised category learning using a Bayesian auto-encoder model, demonstrating improved clustering and categorization performance on image datasets.

Contribution

It introduces a prior encouraging prototype use in a variational auto-encoder for semi-supervised learning, showing benefits over traditional methods.

Findings

Prototypes improve semi-supervised learning performance.

Models form meaningful clustered representations without labels.

Enhanced downstream categorization accuracy.

Abstract

Categories can be represented at different levels of abstraction, from prototypes focused on the most typical members to remembering all observed exemplars of the category. These representations have been explored in the context of supervised learning, where stimuli are presented with known category labels. We examine the benefits of prototype-based representations in a less-studied domain: semi-supervised learning, where agents must form unsupervised representations of stimuli before receiving category labels. We study this problem in a Bayesian unsupervised learning model called a variational auto-encoder, and we draw on recent advances in machine learning to implement a prior that encourages the model to use abstract prototypes to represent data. We apply this approach to image datasets and show that forming prototypes can improve semi-supervised category learning. Additionally, we study the latent embeddings of the models and show that these prototypes allow the models to form clustered representations without supervision, contributing to their success in downstream categorization performance.