Self-Supervised Meta-Learning for Few-Shot Natural Language Classification Tasks

TL;DR

This paper introduces a self-supervised meta-learning approach that generates diverse NLP tasks from unlabeled text to improve few-shot learning, outperforming traditional pre-training and fine-tuning methods.

Contribution

It proposes a novel self-supervised task generation method for meta-learning in NLP, reducing reliance on labeled tasks and enhancing few-shot generalization capabilities.

Findings

Meta-training on generated tasks improves few-shot performance.

Combining self-supervised and supervised tasks yields higher accuracy.

Outperforms standard pre-training followed by fine-tuning on 17 NLP tasks.

Abstract

Self-supervised pre-training of transformer models has revolutionized NLP applications. Such pre-training with language modeling objectives provides a useful initial point for parameters that generalize well to new tasks with fine-tuning. However, fine-tuning is still data inefficient -- when there are few labeled examples, accuracy can be low. Data efficiency can be improved by optimizing pre-training directly for future fine-tuning with few examples; this can be treated as a meta-learning problem. However, standard meta-learning techniques require many training tasks in order to generalize; unfortunately, finding a diverse set of such supervised tasks is usually difficult. This paper proposes a self-supervised approach to generate a large, rich, meta-learning task distribution from unlabeled text. This is achieved using a cloze-style objective, but creating separate multi-class classification tasks by gathering tokens-to-be blanked from among only a handful of vocabulary terms. This yields as many unique meta-training tasks as the number of subsets of vocabulary terms. We meta-train a transformer model on this distribution of tasks using a recent meta-learning framework. On 17 NLP tasks, we show that this meta-training leads to better few-shot generalization than language-model pre-training followed by finetuning. Furthermore, we show how the self-supervised tasks can be combined with supervised tasks for meta-learning, providing substantial accuracy gains over previous supervised meta-learning.