Entropic Causal Inference: Identifiability and Finite Sample Results
This work provides theoretical identifiability guarantees for causal inference methods, which is a foundational problem for researchers and practitioners in causal discovery.
This paper addresses the problem of inferring causal direction between two categorical variables from observational data, proving that the causal direction is identifiable for almost all causal models where the exogenous variable's entropy does not scale with the number of observed variable states. It also provides the first algorithmic identifiability guarantees for the minimum entropy coupling-based approach using a finite number of samples.
Entropic causal inference is a framework for inferring the causal direction between two categorical variables from observational data. The central assumption is that the amount of unobserved randomness in the system is not too large. This unobserved randomness is measured by the entropy of the exogenous variable in the underlying structural causal model, which governs the causal relation between the observed variables. Kocaoglu et al. conjectured that the causal direction is identifiable when the entropy of the exogenous variable is not too large. In this paper, we prove a variant of their conjecture. Namely, we show that for almost all causal models where the exogenous variable has entropy that does not scale with the number of states of the observed variables, the causal direction is identifiable from observational data. We also consider the minimum entropy coupling-based algorithmic approach presented by Kocaoglu et al., and for the first time demonstrate algorithmic identifiability guarantees using a finite number of samples. We conduct extensive experiments to evaluate the robustness of the method to relaxing some of the assumptions in our theory and demonstrate that both the constant-entropy exogenous variable and the no latent confounder assumptions can be relaxed in practice. We also empirically characterize the number of observational samples needed for causal identification. Finally, we apply the algorithm on Tuebingen cause-effect pairs dataset.