A novel hybrid approach for positive-valued DAG learning

arXiv:2604.08935v1 Announce Type: cross Abstract: Causal discovery from observational data remains a fundamental challenge in machine learning and statistics, particularly when variables represent inherently positive quantities such as gene expression levels, asset prices, company revenues, or popul