Logical Distillation of Graph Neural Networks

2024/06/11 by Alexander Pluska, Pascal Welke, Pluska, Alexander +5 · 3 citations
Computer Science · #Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Neural Networks and Applications

paper · pdf · doi:10.48550/arxiv.2406.07126

openalex publication_date 2024/06/11 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We present a logic based interpretable model for learning on graphs and an algorithm to distill this model from a Graph Neural Network (GNN). Recent results have shown connections between the expressivity of GNNs and the two-variable fragment of first-order logic with counting quantifiers (C2). We introduce a decision-tree based model which leverages an extension of C2 to distill interpretable logical classifiers from GNNs. We test our approach on multiple GNN architectures. The distilled models are interpretable, succinct, and attain similar accuracy to the underlying GNN. Furthermore, when the ground truth is expressible in C2, our approach outperforms the GNN.

Cited by

• Sound Logical Explanations for Mean Aggregation Graph Neural Networks
• Discrete Diffusion-Based Model-Level Explanation of Heterogeneous GNNs with Node Features
• WILTing Trees: Interpreting the Distance Between MPNN Embeddings

Related

• Decidability of Graph Neural Networks via Logical Characterizations
• The Logic of Graph Neural Networks
• Calibrate and Boost Logical Expressiveness of GNN Over Multi-Relational and Temporal Graphs
• Can Graph Neural Networks Help Logic Reasoning?
• The Correspondence Between Bounded Graph Neural Networks and Fragments of First-Order Logic