{"ID":6023445,"CreatedAt":"2026-07-08T01:00:23.257252134Z","UpdatedAt":"2026-07-10T08:15:11.905439937Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.05984","arxiv_id":"2607.05984","title":"Learning Sparsest Linear Causal DAGs with Latent Confounders via Higher-Order Cumulants","abstract":"Recovering the exact directed acyclic graph (DAG) in linear non-Gaussian acyclic models with latent confounders (LvLiNGAM) remains a challenging problem. Although LvLiNGAM is identifiable only up to an observational equivalence class, each equivalence class is characterized by a unique sparsest DAG. Recovering the sparsest DAG from finite samples, however, remains difficult. Although existing methods are asymptotically consistent, they do not provide an explicit finite-sample procedure for recovering the unique sparsest DAG, nor do they handle models with an arbitrary number of latent confounders. In this paper, we propose a finite-sample method for recovering the sparsest DAG without imposing any restriction on the number of latent confounders. Simulation studies and real-data analyses demonstrate that the proposed method achieves superior finite-sample performance compared with existing approaches.","short_abstract":"Recovering the exact directed acyclic graph (DAG) in linear non-Gaussian acyclic models with latent confounders (LvLiNGAM) remains a challenging problem. Although LvLiNGAM is identifiable only up to an observational equivalence class, each equivalence class is characterized by a unique sparsest DAG. Recovering the spar...","url_abs":"https://arxiv.org/abs/2607.05984","url_pdf":"https://arxiv.org/pdf/2607.05984v1","authors":"[\"Ming Cai\",\"Hisayuki Hara\"]","published":"2026-07-07T08:17:47Z","proceeding":"cs.LG","tasks":"[\"cs.LG\"]","methods":"[]","has_code":false}