Structurally Separated DAG Learning with Multi-Scale Normalized Closure
Continuous DAG learning commonly enforces acyclicity through scalar functions of a weighted adjacency that aggregate evidence from weighted closed walks. We add a complementary layer: matrix-valued closure logits---explicit multi-hop structural relation scores for each ordered pair---supervised alongside data fit, sparsity, and a DAGMA-style barrier on the weighted adjacency. The distinction is…
We haven't written up this one. arXiv cs.AI has the full story — the link below goes straight to it.
Source: arxiv.org