When Comparing to Ground Truth is Wrong
Authors:
Open Access
- 14 August 2021
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 332-341
Abstract
We study the evaluation of graph explanation methods. The state of the art to evaluate explanation methods is to first train a GNN, then generate explanations, and finally compare those explanations with the ground truth. We show five pitfalls that sabotage this pipeline because the GNN does not use the ground-truth edges. Thus, the explanation method cannot detect the ground truth. We propose three novel benchmarks: (i) pattern detection, (ii) community detection, and (iii) handling negative evidence and gradient saturation. In a re-evaluation of state-of-the-art explanation methods, we show paths for improving existing methods and highlight further paths for GNN explanation research.Keywords
This publication has 7 references indexed in Scilit:
- Explainability Methods for Graph Convolutional Neural Networks2019
- Attention is not not Explanation2019
- Adversarial Attacks on Neural Networks for Graph Data2018
- Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization2017
- "Why Should I Trust You?"2016
- Visualizing and Understanding Convolutional Networks2014
- Structure-activity relationship of mutagenic aromatic and heteroaromatic nitro compounds. Correlation with molecular orbital energies and hydrophobicityJournal of Medicinal Chemistry, 1991