Generic and Trend-aware Curriculum Learning for Relation Extraction in Graph Neural Networks

TL;DR

This paper introduces a generic, trend-aware curriculum learning method for graph neural networks that improves relation extraction by better distinguishing sample difficulty and integrating textual and structural data.

Contribution

It proposes a novel curriculum learning approach that incorporates loss trend analysis to enhance GNN performance in relation extraction tasks.

Findings

Robust estimation of sample difficulty.

Significant performance improvements over state-of-the-art methods.

Effective integration of textual and structural information.

Abstract

We present a generic and trend-aware curriculum learning approach for graph neural networks. It extends existing approaches by incorporating sample-level loss trends to better discriminate easier from harder samples and schedule them for training. The model effectively integrates textual and structural information for relation extraction in text graphs. Experimental results show that the model provides robust estimations of sample difficulty and shows sizable improvement over the state-of-the-art approaches across several datasets.