OrphicX: A Causality-Inspired Latent Variable Model for Interpreting Graph Neural Networks
This addresses the need for interpretable AI in graph-based applications, such as drug discovery, by providing a novel explanation framework, though it is incremental in advancing causal methods for GNNs.
The paper tackles the problem of interpreting graph neural networks (GNNs) by proposing OrphicX, a causality-inspired latent variable model that generates causal explanations, and it shows that OrphicX significantly outperforms alternatives in identifying causal semantics on tasks like molecular graph classification.
This paper proposes a new eXplanation framework, called OrphicX, for generating causal explanations for any graph neural networks (GNNs) based on learned latent causal factors. Specifically, we construct a distinct generative model and design an objective function that encourages the generative model to produce causal, compact, and faithful explanations. This is achieved by isolating the causal factors in the latent space of graphs by maximizing the information flow measurements. We theoretically analyze the cause-effect relationships in the proposed causal graph, identify node attributes as confounders between graphs and GNN predictions, and circumvent such confounder effect by leveraging the backdoor adjustment formula. Our framework is compatible with any GNNs, and it does not require access to the process by which the target GNN produces its predictions. In addition, it does not rely on the linear-independence assumption of the explained features, nor require prior knowledge on the graph learning tasks. We show a proof-of-concept of OrphicX on canonical classification problems on graph data. In particular, we analyze the explanatory subgraphs obtained from explanations for molecular graphs (i.e., Mutag) and quantitatively evaluate the explanation performance with frequently occurring subgraph patterns. Empirically, we show that OrphicX can effectively identify the causal semantics for generating causal explanations, significantly outperforming its alternatives.