Beyond Topological Self-Explainable GNNs: A Formal Explainability Perspective

• URL: http://arxiv.org/abs/2502.02719v1
• Date: Tue, 04 Feb 2025 21:08:23 GMT
• Title: Beyond Topological Self-Explainable GNNs: A Formal Explainability Perspective
• Authors: Steve Azzolin, Sagar Malhotra, Andrea Passerini, Stefano Teso,
• Abstract summary: Self-Explainable Graph Neural Networks (SE-GNNs) are popular explainable-by-design GNNs, but the properties and the limitations of their explanations are not well understood.<n>Our first contribution fills this gap by formalizing the explanations extracted by SE-GNNs, referred to as Trivial Explanations (TEs)<n>We propose Dual-Channel GNNs that integrate a white-box rule extractor and a standard SE-GNN, adaptively combining both channels when the task benefits.
• Score: 19.270404394350944
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Self-Explainable Graph Neural Networks (SE-GNNs) are popular explainable-by-design GNNs, but the properties and the limitations of their explanations are not well understood. Our first contribution fills this gap by formalizing the explanations extracted by SE-GNNs, referred to as Trivial Explanations (TEs), and comparing them to established notions of explanations, namely Prime Implicant (PI) and faithful explanations. Our analysis reveals that TEs match PI explanations for a restricted but significant family of tasks. In general, however, they can be less informative than PI explanations and are surprisingly misaligned with widely accepted notions of faithfulness. Although faithful and PI explanations are informative, they are intractable to find and we show that they can be prohibitively large. Motivated by this, we propose Dual-Channel GNNs that integrate a white-box rule extractor and a standard SE-GNN, adaptively combining both channels when the task benefits. Our experiments show that even a simple instantiation of Dual-Channel GNNs can recover succinct rules and perform on par or better than widely used SE-GNNs. Our code can be found in the supplementary material.

Related papers

• GNN Explanations that do not Explain and How to find Them [20.68967246188274]
  We identify a critical failure of SE-GNN explanations: explanations can be unambiguously unrelated to how the SE-GNNs infer labels.<n>Our empirical analysis reveals that degenerate explanations can be maliciously planted (allowing an attacker to hide the use of sensitive attributes) and can also emerge naturally.<n>To address this, we introduce a novel faithfulness metric that reliably marks degenerate explanations as unfaithful.
  arXiv  Detail & Related papers  (2026-01-28T18:05:17Z)
• Global Graph Counterfactual Explanation: A Subgraph Mapping Approach [54.42907350881448]
  Graph Neural Networks (GNNs) have been widely deployed in various real-world applications. Counterfactual explanation aims to find minimum perturbations on input graphs that change the GNN predictions. We propose GlobalGCE, a novel global-level graph counterfactual explanation method.
  arXiv  Detail & Related papers  (2024-10-25T21:39:05Z)
• Perks and Pitfalls of Faithfulness in Regular, Self-Explainable and Domain Invariant GNNs [18.33293911039292]
  A key desideratum is that explanations are faithful, i.e., that they portray an accurate picture of the GNN's reasoning process. A number of different faithfulness metrics exist, begging the question of what faithfulness is exactly, and what its properties are. We show that, surprisingly, optimizing for faithfulness is not always a sensible design goal.
  arXiv  Detail & Related papers  (2024-06-21T14:01:23Z)
• Explainable Graph Neural Networks Under Fire [69.15708723429307]
  Graph neural networks (GNNs) usually lack interpretability due to their complex computational behavior and the abstract nature of graphs. Most GNN explanation methods work in a post-hoc manner and provide explanations in the form of a small subset of important edges and/or nodes. In this paper we demonstrate that these explanations can unfortunately not be trusted, as common GNN explanation methods turn out to be highly susceptible to adversarial perturbations.
  arXiv  Detail & Related papers  (2024-06-10T16:09:16Z)
• Graph Neural Network Explanations are Fragile [39.9987140075811]
  We take the first step to study GNN explainers under adversarial attack. We find that an adversary slightly perturbing graph structure can ensure GNN model makes correct predictions, but the GNN explainer yields a drastically different explanation on the perturbed graph.
  arXiv  Detail & Related papers  (2024-06-05T12:23:02Z)
• Understanding Expressivity of GNN in Rule Learning [36.04983130825589]
  Rule learning is critical to improving knowledge graph (KG) reasoning. GNNs with tail entity scoring are unified into a common framework. We propose a novel labeling strategy to learn more rules in KG reasoning.
  arXiv  Detail & Related papers  (2023-03-22T04:49:00Z)
• On Consistency in Graph Neural Network Interpretation [34.25952902469481]
  Instance-level GNN explanation aims to discover critical input elements, like nodes or edges, that the target GNN relies upon for making predictions. Various algorithms are proposed, but most of them formalize this task by searching the minimal subgraph. We propose a simple yet effective countermeasure by aligning embeddings.
  arXiv  Detail & Related papers  (2022-05-27T02:58:07Z)
• Task-Agnostic Graph Explanations [50.17442349253348]
  Graph Neural Networks (GNNs) have emerged as powerful tools to encode graph structured data. Existing learning-based GNN explanation approaches are task-specific in training. We propose a Task-Agnostic GNN Explainer (TAGE) trained under self-supervision with no knowledge of downstream tasks.
  arXiv  Detail & Related papers  (2022-02-16T21:11:47Z)
• ProtGNN: Towards Self-Explaining Graph Neural Networks [12.789013658551454]
  We propose Prototype Graph Neural Network (ProtGNN), which combines prototype learning with GNNs. ProtGNN and ProtGNN+ can provide inherent interpretability while achieving accuracy on par with the non-interpretable counterparts.
  arXiv  Detail & Related papers  (2021-12-02T01:16:29Z)
• Towards Self-Explainable Graph Neural Network [24.18369781999988]
  Graph Neural Networks (GNNs) generalize the deep neural networks to graph-structured data. GNNs lack explainability, which limits their adoption in scenarios that demand the transparency of models. We propose a new framework which can find $K$-nearest labeled nodes for each unlabeled node to give explainable node classification.
  arXiv  Detail & Related papers  (2021-08-26T22:45:11Z)
• Jointly Attacking Graph Neural Network and its Explanations [50.231829335996814]
  Graph Neural Networks (GNNs) have boosted the performance for many graph-related tasks. Recent studies have shown that GNNs are highly vulnerable to adversarial attacks, where adversaries can mislead the GNNs' prediction by modifying graphs. We propose a novel attack framework (GEAttack) which can attack both a GNN model and its explanations by simultaneously exploiting their vulnerabilities.
  arXiv  Detail & Related papers  (2021-08-07T07:44:33Z)
• Optimization of Graph Neural Networks: Implicit Acceleration by Skip Connections and More Depth [57.10183643449905]
  Graph Neural Networks (GNNs) have been studied from the lens of expressive power and generalization. We study the dynamics of GNNs by studying deep skip optimization. Our results provide first theoretical support for the success of GNNs.
  arXiv  Detail & Related papers  (2021-05-10T17:59:01Z)
• Parameterized Explainer for Graph Neural Network [49.79917262156429]
  We propose PGExplainer, a parameterized explainer for Graph Neural Networks (GNNs) Compared to the existing work, PGExplainer has better generalization ability and can be utilized in an inductive setting easily. Experiments on both synthetic and real-life datasets show highly competitive performance with up to 24.7% relative improvement in AUC on explaining graph classification.
  arXiv  Detail & Related papers  (2020-11-09T17:15:03Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.