Deep neural networks (DNNs) achieve state-of-the-art performance on many tasks, but this often requires increasingly larger model sizes, which in turn leads to more complex internal representations. Explainability techniques (XAI) have made remarkable progress in the interpretability of ML models. However, the non-relational nature of Graph neural networks (GNNs) make it difficult to reuse already existing XAI methods. While other works have focused on instance-based explanation methods for GNNs, very few have investigated model-based methods and, to our knowledge, none have tried to probe the embedding of the GNNs for well-known structural graph properties. In this paper we present a model agnostic explainability pipeline for GNNs employing diagnostic classifiers. This pipeline aims to probe and interpret the learned representations in GNNs across various architectures and datasets, refining our understanding and trust in these models.

翻译：深度神经网络（DNNs）在许多任务上实现了最先进的性能，但这通常需要越来越大的模型规模，进而导致更复杂的内部表示。可解释性技术（XAI）在机器学习模型的可解释性方面取得了显著进展。然而，图神经网络（GNNs）的非关系特性使得难以直接复用现有的XAI方法。虽然已有研究聚焦于GNNs的实例级解释方法，但极少有工作探索模型级方法，且据我们所知，尚未有研究尝试从GNNs的嵌入表示中探测已知的图结构属性。本文提出一种采用诊断分类器的模型无关GNN可解释性分析框架。该框架旨在跨不同架构和数据集探测并解释GNNs学习到的表征，从而深化我们对这些模型的理解与信任。