Truncated Matrix Power Iteration for Differentiable DAG Learning. (arXiv:2208.14571v1 [cs.LG])

Recovering underlying Directed Acyclic Graph structures (DAG) from
observational data is highly challenging due to the combinatorial nature of the
DAG-constrained optimization problem. Recently, DAG learning has been cast as a
continuous optimization problem by characterizing the DAG constraint as a
smooth equality one, generally based on polynomials over adjacency matrices.
Existing methods place very small coefficients on high-order polynomial terms
for stabilization, since they argue that large coefficients on the higher-order
terms are harmful due to numeric exploding. On …

arxiv dag iteration learning power