Discovering Invariant Rationales for Graph Neural Networks

By Yingmin Wu, Xiang Wang, An Zhang, Xiangnan He, Tat-seng Chua. International Conference on Learning Representations, 2022

326 citations recorded by the source corpus.

Intrinsic interpretability of graph neural networks (GNNs) is to find a small subset of the input graph's features -- rationale -- which guides the model prediction. Unfortunately, the leading rationalization models often rely on data biases, especially shortcut features, to compose rationales and make predictions without probing the critical and causal patterns. Moreover, such data biases easily change outside the training distribution. As a result, these models suffer from a huge drop in interpretability and predictive performance on out-of-distribution data. In this work, we propose a new strategy of discovering invariant rationale (DIR) to construct intrinsically interpretable GNNs. It conducts interventions on the training distribution to create multiple interventional distributions. Then it approaches the causal rationales that are invariant across different distributions while filtering out the spurious patterns that are unstable. Experiments on both synthetic and real-world datasets validate the superiority of our DIR in terms of interpretability and generalization ability on graph classification over the leading baselines. Code and datasets are available at https://github.com/Wuyxin/DIR-GNN.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/bd15a322c20f891f38e247bd5ed6e9d2f0b637eb

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

  1. Siqi Miao
  2. Miaoyuan Liu
  3. Pan Li

Citations in the corpus