Meta-learning Pathologies from Radiology Reports using Variance Aware Prototypical Networks

TL;DR

This paper introduces a Gaussian-based extension of Prototypical Networks for few-shot text classification, improving performance and enabling out-of-distribution detection in radiology report analysis.

Contribution

It proposes a novel Gaussian prototype approach with regularization for few-shot NLP tasks, enhancing classification accuracy and OOD detection capabilities.

Findings

Outperforms strong baselines on 17 datasets

Effective in detecting out-of-distribution samples

Improves few-shot classification accuracy

Abstract

Large pretrained Transformer-based language models like BERT and GPT have changed the landscape of Natural Language Processing (NLP). However, fine tuning such models still requires a large number of training examples for each target task, thus annotating multiple datasets and training these models on various downstream tasks becomes time consuming and expensive. In this work, we propose a simple extension of the Prototypical Networks for few-shot text classification. Our main idea is to replace the class prototypes by Gaussians and introduce a regularization term that encourages the examples to be clustered near the appropriate class centroids. Experimental results show that our method outperforms various strong baselines on 13 public and 4 internal datasets. Furthermore, we use the class distributions as a tool for detecting potential out-of-distribution (OOD) data points during deployment.