In linear non-Gaussian models, high-dimensional cyclic causal structures can be coarsened to a low-dimensional DAG that is invariant across observational equivalence classes and learnable in cubic time.
Maeda and Shohei Shimizu
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Coarsening Linear Non-Gaussian Causal Models with Cycles
In linear non-Gaussian models, high-dimensional cyclic causal structures can be coarsened to a low-dimensional DAG that is invariant across observational equivalence classes and learnable in cubic time.