Causal normalizing flows: from theory to practice
This work addresses causal inference for researchers and practitioners dealing with complex, mixed-type data and partial graph knowledge, offering a novel method but with incremental improvements over existing techniques.
The paper tackles the problem of causal reasoning from observational data by proposing causal normalizing flows, showing they can recover causal models and answer interventional and counterfactual questions, with validation through experiments including comparisons to other methods and application to real-world mixed data scenarios.
In this work, we deepen on the use of normalizing flows for causal reasoning. Specifically, we first leverage recent results on non-linear ICA to show that causal models are identifiable from observational data given a causal ordering, and thus can be recovered using autoregressive normalizing flows (NFs). Second, we analyze different design and learning choices for causal normalizing flows to capture the underlying causal data-generating process. Third, we describe how to implement the do-operator in causal NFs, and thus, how to answer interventional and counterfactual questions. Finally, in our experiments, we validate our design and training choices through a comprehensive ablation study; compare causal NFs to other approaches for approximating causal models; and empirically demonstrate that causal NFs can be used to address real-world problems, where the presence of mixed discrete-continuous data and partial knowledge on the causal graph is the norm. The code for this work can be found at https://github.com/psanch21/causal-flows.