MICK: A Meta-Learning Framework for Few-shot Relation Classification with Small Training Data
This addresses the problem of few-shot learning in relation classification for domains like health where annotated data is scarce, though it is incremental in improving existing methods.
The paper tackles few-shot relation classification with very limited training data by proposing a meta-learning framework that enhances instance representations and aggregates cross-domain knowledge, achieving state-of-the-art results with small data and competitive performance with larger data.
Few-shot relation classification seeks to classify incoming query instances after meeting only few support instances. This ability is gained by training with large amount of in-domain annotated data. In this paper, we tackle an even harder problem by further limiting the amount of data available at training time. We propose a few-shot learning framework for relation classification, which is particularly powerful when the training data is very small. In this framework, models not only strive to classify query instances, but also seek underlying knowledge about the support instances to obtain better instance representations. The framework also includes a method for aggregating cross-domain knowledge into models by open-source task enrichment. Additionally, we construct a brand new dataset: the TinyRel-CM dataset, a few-shot relation classification dataset in health domain with purposely small training data and challenging relation classes. Experimental results demonstrate that our framework brings performance gains for most underlying classification models, outperforms the state-of-the-art results given small training data, and achieves competitive results with sufficiently large training data.