GraphNarrator: Generating Textual Explanations for Graph Neural Networks

TL;DR

GraphNarrator introduces a novel method that generates natural language explanations for Graph Neural Networks by leveraging pseudo-labels and iterative training, enhancing interpretability in graph-based models.

Contribution

It is the first approach to produce textual explanations for GNNs, combining pseudo-label generation with expert iteration for improved explanation quality.

Findings

Produces faithful and human-preferred explanations

Effective in various graph learning applications

Outperforms baseline explanation methods

Abstract

Graph representation learning has garnered significant attention due to its broad applications in various domains, such as recommendation systems and social network analysis. Despite advancements in graph learning methods, challenges still remain in explainability when graphs are associated with semantic features. In this paper, we present GraphNarrator, the first method designed to generate natural language explanations for Graph Neural Networks. GraphNarrator employs a generative language model that maps input-output pairs to explanations reflecting the model's decision-making process. To address the lack of ground truth explanations to train the model, we propose first generating pseudo-labels that capture the model's decisions from saliency-based explanations, then using Expert Iteration to iteratively train the pseudo-label generator based on training objectives on explanation quality. The high-quality pseudo-labels are finally utilized to train an end-to-end explanation generator model. Extensive experiments are conducted to demonstrate the effectiveness of GraphNarrator in producing faithful, concise, and human-preferred natural language explanations.