Published on Wed Apr 14 2021

Generative Causal Explanations for Graph Neural Networks

Wanyu Lin, Hao Lan, Baochun Li
0
0
0
Abstract

This paper presents Gem, a model-agnostic approach for providing interpretable explanations for any GNNs on various graph learning tasks. Specifically, we formulate the problem of providing explanations for the decisions of GNNs as a causal learning task. Then we train a causal explanation model equipped with a loss function based on Granger causality. Different from existing explainers for GNNs, Gem explains GNNs on graph-structured data from a causal perspective. It has better generalization ability as it has no requirements on the internal structure of the GNNs or prior knowledge on the graph learning tasks. In addition, Gem, once trained, can be used to explain the target GNN very quickly. Our theoretical analysis shows that several recent explainers fall into a unified framework of additive feature attribution methods. Experimental results on synthetic and real-world datasets show that Gem achieves a relative increase of the explanation accuracy by up to and speeds up the explanation process by up to as compared to its state-of-the-art alternatives.

Sun Mar 10 2019
Machine Learning
GNNExplainer: Generating Explanations for Graph Neural Networks
Graph Neural Networks (GNNs) are a powerful tool for machine learning on graphs. GNNExplainer is the first general, model-agnostic approach for providing explanations for predictions of any GNN-based model.
0
0
0
Mon Oct 12 2020
Machine Learning
PGM-Explainer: Probabilistic Graphical Model Explanations for Graph Neural Networks
PGM-Explainer is a Probabilistic Graphical Model (PGM) model-agnostic explainer for GNNs. Given a prediction to be explained, PGM-explainer identifies crucial graph components and generates an explanation in form of a PGM approximating that prediction.
0
0
0
Fri Jan 17 2020
Machine Learning
GraphLIME: Local Interpretable Model Explanations for Graph Neural Networks
Graph structured data has wide applicability in various domains such as physics, chemistry, biology, computer vision, and social networks. GraphLIME is a generic GNN-model explanation framework that learns a nonlinear interpretable model locally in the subgraph being explained.
0
0
0
Wed Jun 03 2020
Machine Learning
XGNN: Towards Model-Level Explanations of Graph Neural Networks
Graphs neural networks (GNNs) learn node features by aggregating and combining neighbor information. GNNs are mostly treated as black-boxes and lack human intelligible explanations. In this work, we propose a novel approach, known as XGNN, to interpret the model-
0
0
0
Mon Nov 09 2020
Artificial Intelligence
Parameterized Explainer for Graph Neural Network
PGExplainer is a parameterized explainer for Graph Neural Networks (GNNs) It uses a deep neural network to parameterize the generation process of explanations. PGExplainer has better generalization ability and can be used in an inductive setting easily.
2
0
1
Tue May 18 2021
Machine Learning
Zorro: Valid, Sparse, and Stable Explanations in Graph Neural Networks
Explanations for a GNN model differ in principle from other input settings. It is important to attribute the decision to inputfeatures and other related instances connected by the graph structure. We propose a novel approach Zorro based on the principles from rate-distortion theory.
1
0
0