Scholarlink

<p>Counterfactual explanations emerged as a powerful tool to unveil the opaque decision-making processes of Graph Neural Networks (GNNs). However, existing techniques primarily focus on edge modifications, often overlooking the crucial role of node features and edge attributes changes in shaping the model predictions. To address this limitation, we propose a novel Unified Counterfactual Explainer (UCExplainer) for GNNs that generates counterfactual explanations for both node and graph classification tasks. Unlike prior methods, UCExplainer generates realistic and interpretable counterfactuals through a unified approach. It balances modifications by jointly optimizing perturbations across edges, edge attributes, and node features, ensuring the minimal changes required to flip a model’s prediction. Furthermore, UCExplainer offers maximum flexibility by supporting: (1) perturbations of both discrete and continuous values, (2) the clipping of values to a user-defined ranges, and (3) the exclusion of specific elements, such as entire nodes, edges, or individual features, from modification. Experiments on real-world datasets demonstrate the effectiveness and robustness of our approach over existing baselines.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Unified counterfactual explainer for graph neural networks

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月4日

查看全文
  • Flavio Giorgi,
  • Fabrizio Silvestri,
  • Gabriele Tolomei

摘要

Counterfactual explanations emerged as a powerful tool to unveil the opaque decision-making processes of Graph Neural Networks (GNNs). However, existing techniques primarily focus on edge modifications, often overlooking the crucial role of node features and edge attributes changes in shaping the model predictions. To address this limitation, we propose a novel Unified Counterfactual Explainer (UCExplainer) for GNNs that generates counterfactual explanations for both node and graph classification tasks. Unlike prior methods, UCExplainer generates realistic and interpretable counterfactuals through a unified approach. It balances modifications by jointly optimizing perturbations across edges, edge attributes, and node features, ensuring the minimal changes required to flip a model’s prediction. Furthermore, UCExplainer offers maximum flexibility by supporting: (1) perturbations of both discrete and continuous values, (2) the clipping of values to a user-defined ranges, and (3) the exclusion of specific elements, such as entire nodes, edges, or individual features, from modification. Experiments on real-world datasets demonstrate the effectiveness and robustness of our approach over existing baselines.