Adversarial Learning for Supervised and Semi-supervised Relation Extraction in Biomedical Literature

TL;DR

This paper explores adversarial training techniques to enhance relation extraction in biomedical literature, demonstrating improvements in supervised and semi-supervised settings and achieving state-of-the-art results.

Contribution

It introduces the use of multiple adversarial examples and semi-supervised adversarial training for relation extraction in biomedical texts, a novel approach in this domain.

Findings

Adversarial training improves supervised relation extraction performance.

Semi-supervised adversarial training effectively utilizes unlabeled data.

The method achieves state-of-the-art results on benchmark datasets.

Abstract

Adversarial training is a technique of improving model performance by involving adversarial examples in the training process. In this paper, we investigate adversarial training with multiple adversarial examples to benefit the relation extraction task. We also apply adversarial training technique in semi-supervised scenarios to utilize unlabeled data. The evaluation results on protein-protein interaction and protein subcellular localization task illustrate adversarial training provides improvement on the supervised model, and is also effective on involving unlabeled data in the semi-supervised training case. In addition, our method achieves state-of-the-art performance on two benchmarking datasets.