Improving One-Shot Learning through Fusing Side Information

TL;DR

This paper presents two statistical methods for integrating side information into deep neural networks to enhance one-shot learning, demonstrating improved performance over existing models.

Contribution

It introduces novel statistical approaches and an attention mechanism to effectively fuse side information for better one-shot learning performance.

Findings

Improved recognition accuracy on one-shot tasks.

Effective use of side information to compensate for limited data.

Outperforms state-of-the-art attentional networks.

Abstract

Deep Neural Networks (DNNs) often struggle with one-shot learning where we have only one or a few labeled training examples per category. In this paper, we argue that by using side information, we may compensate the missing information across classes. We introduce two statistical approaches for fusing side information into data representation learning to improve one-shot learning. First, we propose to enforce the statistical dependency between data representations and multiple types of side information. Second, we introduce an attention mechanism to efficiently treat examples belonging to the 'lots-of-examples' classes as quasi-samples (additional training samples) for 'one-example' classes. We empirically show that our learning architecture improves over traditional softmax regression networks as well as state-of-the-art attentional regression networks on one-shot recognition tasks.