ProtGNN: Towards Self-Explaining Graph Neural Networks

Zaixi Zhang, Qi Liu, Hao Wang, Chengqiang Lu, Chee‐Kong Lee

Proceedings of the AAAI Conference on Artificial Intelligence · 2022 · 106 citations · 38 references

DOIFull text

Open access

Concepts

TL;DR

Despite recent advances in GNNs, explaining their predictions remains difficult because post‑hoc methods cannot reveal the model’s original reasoning, underscoring the need for inherently interpretable GNNs. This work introduces ProtGNN, a prototype‑based GNN that embeds interpretability directly into the model’s architecture. ProtGNN derives explanations from a case‑based reasoning process by comparing inputs to learned latent‑space prototypes, and its enhanced ProtGNN+ variant uses a conditional subgraph sampling module to highlight the graph regions most similar to each prototype during classification. Experiments across diverse datasets show that ProtGNN and ProtGNN+ achieve accuracy comparable to black‑box GNNs while offering inherent, interpretable explanations.

Abstract

Despite the recent progress in Graph Neural Networks (GNNs), it remains challenging to explain the predictions made by GNNs. Existing explanation methods mainly focus on post-hoc explanations where another explanatory model is employed to provide explanations for a trained GNN. The fact that post-hoc methods fail to reveal the original reasoning process of GNNs raises the need of building GNNs with built-in interpretability. In this work, we propose Prototype Graph Neural Network (ProtGNN), which combines prototype learning with GNNs and provides a new perspective on the explanations of GNNs. In ProtGNN, the explanations are naturally derived from the case-based reasoning process and are actually used during classification. The prediction of ProtGNN is obtained by comparing the inputs to a few learned prototypes in the latent space. Furthermore, for better interpretability and higher efficiency, a novel conditional subgraph sampling module is incorporated to indicate which part of the input graph is most similar to each prototype in ProtGNN+. Finally, we evaluate our method on a wide range of datasets and perform concrete case studies. Extensive results show that ProtGNN and ProtGNN+ can provide inherent interpretability while achieving accuracy on par with the non-interpretable counterparts.

References

38