PISA: Prioritized Invariant Subgraph Aggregation
By: Ali Ghasemi , Farooq Ahmad Wani , Maria Sofia Bucarelli and more
Potential Business Impact:
Finds better patterns in complex data for more accurate predictions.
Recent work has extended the invariance principle for out-of-distribution (OOD) generalization from Euclidean to graph data, where challenges arise due to complex structures and diverse distribution shifts in node attributes and topology. To handle these, Chen et al. proposed CIGA (Chen et al., 2022b), which uses causal modeling and an information-theoretic objective to extract a single invariant subgraph capturing causal features. However, this single-subgraph focus can miss multiple causal patterns. Liu et al. (2025) addressed this with SuGAr, which learns and aggregates diverse invariant subgraphs via a sampler and diversity regularizer, improving robustness but still relying on simple uniform or greedy aggregation. To overcome this, the proposed PISA framework introduces a dynamic MLP-based aggregation that prioritizes and combines subgraph representations more effectively. Experiments on 15 datasets, including DrugOOD (Ji et al., 2023), show that PISA achieves up to 5% higher classification accuracy than prior methods.
Similar Papers
Learning Invariant Graph Representations Through Redundant Information
Machine Learning (CS)
Helps computers learn from new data without mistakes.
On Conditional Independence Graph Learning From Multi-Attribute Gaussian Dependent Time Series
Machine Learning (Stat)
Helps computers understand complex data relationships.
Invariant Graph Transformer for Out-of-Distribution Generalization
Machine Learning (CS)
Helps computers understand different kinds of graphs.