Lifted Causal Inference in Relational Domains
This enables more efficient causal analysis in relational data, which is incremental as it extends existing lifting techniques to causal inference.
The paper tackles the problem of inefficient causal inference in relational domains by applying lifting techniques from probabilistic inference, resulting in drastically faster computation of causal effects compared to propositional methods.
Lifted inference exploits symmetries in probabilistic graphical models by using a representative for indistinguishable objects, thereby speeding up query answering while maintaining exact answers. Even though lifting is a well-established technique for the task of probabilistic inference in relational domains, it has not yet been applied to the task of causal inference. In this paper, we show how lifting can be applied to efficiently compute causal effects in relational domains. More specifically, we introduce parametric causal factor graphs as an extension of parametric factor graphs incorporating causal knowledge and give a formal semantics of interventions therein. We further present the lifted causal inference algorithm to compute causal effects on a lifted level, thereby drastically speeding up causal inference compared to propositional inference, e.g., in causal Bayesian networks. In our empirical evaluation, we demonstrate the effectiveness of our approach.