Optimal computational complexity for learning interventional Markov equivalence classes

Determine the optimal computational complexity for learning interventional Markov equivalence classes under the causal-discovery setting considered in the paper.

Background

The paper contrasts its constraint-query results with broader computational-complexity questions for interventional Markov equivalence classes (IMECs). It establishes constraint-query optimality up to a multiplicative constant for maximal ancestral graphs under the specified intervention framework, but distinguishes this result from the optimal overall computational complexity of learning an IMEC. The latter complexity remains unresolved.

References

However, the optimal computational complexity for IMEC remains unknown.

— Statistical Inference for Causal Discovery under Selection and Latent Variables via Single-Target Interventions  (2609.28856 - Hou et al., 23 Sep 2026) in Section 1, Related Work