Causal Lifting and Link Prediction
This addresses a bottleneck in causal inference for graph-based tasks like recommendations, offering a novel approach for path-dependent scenarios, though it is incremental in extending causal modeling to specific graph contexts.
The paper tackles the problem of path-dependent link formation in causal models for link prediction, where existing methods fail due to unidentifiable dependencies or impractical control requirements. It introduces causal lifting to enable identification with limited interventional data and shows that structural pairwise embeddings reduce bias and better represent causal structure, validated on tasks like knowledge base completion and recommendations.
Existing causal models for link prediction assume an underlying set of inherent node factors -- an innate characteristic defined at the node's birth -- that governs the causal evolution of links in the graph. In some causal tasks, however, link formation is path-dependent: The outcome of link interventions depends on existing links. Unfortunately, these existing causal methods are not designed for path-dependent link formation, as the cascading functional dependencies between links (arising from path dependence) are either unidentifiable or require an impractical number of control variables. To overcome this, we develop the first causal model capable of dealing with path dependencies in link prediction. In this work we introduce the concept of causal lifting, an invariance in causal models of independent interest that, on graphs, allows the identification of causal link prediction queries using limited interventional data. Further, we show how structural pairwise embeddings exhibit lower bias and correctly represent the task's causal structure, as opposed to existing node embeddings, e.g., graph neural network node embeddings and matrix factorization. Finally, we validate our theoretical findings on three scenarios for causal link prediction tasks: knowledge base completion, covariance matrix estimation and consumer-product recommendations.