グラフ注意に基づく説明の意味的解釈と検証
Semantic Interpretation and Validation of Graph Attention-based Explanations for GNN Models
GNNモデルの説明可能性を高めるため、注意重みを特徴量重要度として使う妥当性を、意味的に整理した特徴量摂動とモデル精度の相関で検証する手法を提案し、LiDAR点群推定モデルで有効性を示した。
著者: Efimia Panagiotaki, Daniele De Martini, Lars Kunze
分類: cs.LG, cs.AI, cs.CY, cs.RO
原文アブストラクト
In this work, we propose a methodology for investigating the use of semantic attention to enhance the explainability of Graph Neural Network (GNN)-based models. Graph Deep Learning (GDL) has emerged as a promising field for tasks like scene interpretation, leveraging flexible graph structures to concisely describe complex features and relationships. As traditional explainability methods used in eXplainable AI (XAI) cannot be directly applied to such structures, graph-specific approaches are introduced. Attention has been previously employed to estimate the importance of input features in GDL, however, the fidelity of this method in generating accurate and consistent explanations has been questioned. To evaluate the validity of using attention weights as feature importance indicators, we introduce semantically-informed perturbations and correlate predicted attention weights with the accuracy of the model. Our work extends existing attention-based graph explainability methods by analysing the divergence in the attention distributions in relation to semantically sorted feature sets and the behaviour of a GNN model, efficiently estimating feature importance. We apply our methodology on a lidar pointcloud estimation model successfully identifying key semantic classes that contribute to enhanced performance, effectively generating reliable post-hoc semantic explanations.
