Robust Causal Graph Representation Learning against Confounding Effects

The prevailing graph neural network models have achieved significant progress in graph representation learning. However, in this paper, we uncover an ever-overlooked phenomenon: the pre-trained graph representation learning model tested with full graphs underperforms the model tested with well-pruned graphs. This observation reveals that there exist confounders in graphs, which may interfere with the model learning semantic information, and current graph representation learning methods have not eliminated their influence. To tackle this issue, we propose Robust Causal Graph Representation Learning (RCGRL) to learn robust graph representations against confounding effects. RCGRL introduces an active approach to generate instrumental variables under unconditional moment restrictions, which empowers the graph representation learning model to eliminate confounders, thereby capturing discriminative information that is causally related to downstream predictions. We offer theorems and proofs to guarantee the theoretical effectiveness of the proposed approach. Empirically, we conduct extensive experiments on a synthetic dataset and multiple benchmark datasets. The results demonstrate that compared with state-of-the-art methods, RCGRL achieves better prediction performance and generalization ability.

[1]  Shuiwang Ji,et al.  Explainability in Graph Neural Networks: A Taxonomic Survey , 2020, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  Xiangnan He,et al.  Discovering Invariant Rationales for Graph Neural Networks , 2022, ICLR.

[3]  Shuiwang Ji,et al.  Graph U-Nets , 2019, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[4]  Baochun Li,et al.  Generative Causal Explanations for Graph Neural Networks , 2021, ICML.

[5]  Jure Leskovec,et al.  OGB-LSC: A Large-Scale Challenge for Machine Learning on Graphs , 2021, NeurIPS Datasets and Benchmarks.

[6]  Shuiwang Ji,et al.  On Explainability of Graph Neural Networks via Subgraph Explorations , 2021, ICML.

[7]  Aaron C. Courville,et al.  Out-of-Distribution Generalization via Risk Extrapolation (REx) , 2020, ICML.

[8]  Bo Zong,et al.  Parameterized Explainer for Graph Neural Network , 2020, NeurIPS.

[9]  J. Leskovec,et al.  Open Graph Benchmark: Datasets for Machine Learning on Graphs , 2020, NeurIPS.

[10]  Tatsunori B. Hashimoto,et al.  Distributionally Robust Neural Networks , 2020, ICLR.

[11]  Yichen Wei,et al.  Circle Loss: A Unified Perspective of Pair Similarity Optimization , 2020, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[12]  P. Talukdar,et al.  ASAP: Adaptive Structure Aware Pooling for Learning Hierarchical Graph Representations , 2019, AAAI.

[13]  Aaron C. Courville,et al.  Out-of-Distribution Generalization via Risk Extrapolation (REx) , 2020, International Conference on Machine Learning.

[14]  David Lopez-Paz,et al.  Invariant Risk Minimization , 2019, ArXiv.

[15]  Jaewoo Kang,et al.  Self-Attention Graph Pooling , 2019, ICML.

[16]  J. Leskovec,et al.  GNNExplainer: Generating Explanations for Graph Neural Networks , 2019, NeurIPS.

[17]  Martin Grohe,et al.  Weisfeiler and Leman Go Neural: Higher-order Graph Neural Networks , 2018, AAAI.

[18]  Jure Leskovec,et al.  How Powerful are Graph Neural Networks? , 2018, ICLR.

[19]  Pietro Liò,et al.  Graph Attention Networks , 2017, ICLR.

[20]  Max Welling,et al.  Semi-Supervised Classification with Graph Convolutional Networks , 2016, ICLR.

[21]  J. Pearl,et al.  Causal Inference in Statistics: A Primer , 2016 .

[22]  L. Hansen,et al.  Finite Sample Properties of Some Alternative Gmm Estimators , 2015 .

[23]  Hong Li,et al.  U-Processes and Preference Learning , 2014, Neural Computation.

[24]  Xiaohong Chen,et al.  Efficient Estimation of Models with Conditional Moment Restrictions Containing Unknown Functions , 2003 .

[25]  Jeffrey M. Woodbridge Econometric Analysis of Cross Section and Panel Data , 2002 .