期刊
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
卷 44, 期 11, 页码 7581-7596出版社
IEEE COMPUTER SOC
DOI: 10.1109/TPAMI.2021.3115452
关键词
Graph neural networks; Neural networks; Predictive models; Optimization; Taylor series; Feature extraction; Adaptation models; Graph neural networks; higher-order explanations; layer-wise relevance propagation; explainable machine learning
资金
- German Ministry for Education and Research [01IS18025A, 01IS18037A]
- German Research Foundation (DFG) as Math+: Berlin Mathematics Research Center [EXC 2046/1, 390685689]
- Institute of Information & Communications Technology Planning & Evaluation (IITP) - Korea Government [2019-0-00079]
This paper introduces a new method for explaining graph neural networks, which can extract relevant input graph traversals that contribute to the prediction using higher-order expansions and nested attribution scheme. It has practical applications in areas such as sentiment analysis of text data, structure-property relationships in quantum chemistry, and image classification.
Graph Neural Networks (GNNs) are a popular approach for predicting graph structured data. As GNNs tightly entangle the input graph into the neural network structure, common explainable AI approaches are not applicable. To a large extent, GNNs have remained black-boxes for the user so far. In this paper, we show that GNNs can in fact be naturally explained using higher-order expansions, i.e., by identifying groups of edges that jointly contribute to the prediction. Practically, we find that such explanations can be extracted using a nested attribution scheme, where existing techniques such as layer-wise relevance propagation (LRP) can be applied at each step. The output is a collection of walks into the input graph that are relevant for the prediction. Our novel explanation method, which we denote by GNN-LRP, is applicable to a broad range of graph neural networks and lets us extract practically relevant insights on sentiment analysis of text data, structure-property relationships in quantum chemistry, and image classification.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据