Faithful and Consistent Graph Neural Network Explanations with Rationale Alignment

TL;DR

This paper introduces a causality-inspired framework for explaining graph neural networks, addressing spurious explanations and improving faithfulness and consistency through an auxiliary alignment loss.

Contribution

It proposes a novel explanation method with an alignment loss, theoretically proven to enhance explanation faithfulness and consistency in GNNs.

Findings

The framework improves explanation faithfulness and consistency.

Different alignment strategies are effective in various scenarios.

Theoretical analysis supports the proposed alignment loss benefits.

Abstract

Uncovering rationales behind predictions of graph neural networks (GNNs) has received increasing attention over recent years. Instance-level GNN explanation aims to discover critical input elements, like nodes or edges, that the target GNN relies upon for making predictions. %These identified sub-structures can provide interpretations of GNN's behavior. Though various algorithms are proposed, most of them formalize this task by searching the minimal subgraph which can preserve original predictions. However, an inductive bias is deep-rooted in this framework: several subgraphs can result in the same or similar outputs as the original graphs. Consequently, they have the danger of providing spurious explanations and failing to provide consistent explanations. Applying them to explain weakly-performed GNNs would further amplify these issues. To address this problem, we theoretically examine the predictions of GNNs from the causality perspective. Two typical reasons for spurious explanations are identified: confounding effect of latent variables like distribution shift, and causal factors distinct from the original input. Observing that both confounding effects and diverse causal rationales are encoded in internal representations, \tianxiang{we propose a new explanation framework with an auxiliary alignment loss, which is theoretically proven to be optimizing a more faithful explanation objective intrinsically. Concretely for this alignment loss, a set of different perspectives are explored: anchor-based alignment, distributional alignment based on Gaussian mixture models, mutual-information-based alignment, etc. A comprehensive study is conducted both on the effectiveness of this new framework in terms of explanation faithfulness/consistency and on the advantages of these variants.