Soft causal learning for generalized molecule property prediction: An environment modeling perspective

Learning on molecule graphs has become an increasingly important topic in AI for science, which takes full advantage of AI to facilitate scientific discovery. Existing solutions on modeling molecules utilize Graph Neural Networks (GNNs) to achieve representations, but they mostly fail to adapt models to out-of-distribution (OOD) samples. Although recent advances on OOD-oriented graph learning have discovered the invariant rationale on graphs, they still ignore three important issues, i.e., 1) the expanding atom patterns regarding environments on graphs lead to failures of invariant rationale-based models, 2) the associations between discovered molecular subgraphs and corresponding properties are complex where causal substructures cannot fully interpret the labels, and 3) the interactions between environments and invariances can influence with each other and thus are challenging to be modeled. To this end, we propose a soft causal learning framework, to tackle the unresolved molecule OOD challenge, from the perspective of negatively modeling the molecule environments and bypassing the invariant subgraphs. Specifically, we first incorporate chemistry theories into our graph growth generator to imitate expanded environments and then devise a GIB-based objective to disentangle environment from whole graphs and finally introduce a cross-attention-based soft causal interaction, which allows dynamic interactions between environments and invariances. We perform extensive experiments on seven datasets by imitating different kinds of OOD generalization scenarios. Extensive comparison, ablation experiments as well as visualized case studies demonstrate well generalization ability of our proposal.