{"ID":5937613,"CreatedAt":"2026-07-07T03:14:33.014478982Z","UpdatedAt":"2026-07-08T06:13:44.489501299Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.04133","arxiv_id":"2607.04133","title":"MDL Meets Latent Confounders: LNML-based Causal Discovery","abstract":"Causal discovery with nonlinear mechanisms and latent confounders remains challenging. Existing methods often rely on either linear assumptions or causal sufficiency, limiting their applicability. We propose an MDL-based causal discovery framework that explicitly accounts for latent confounders while allowing flexible nonlinear mechanisms by minimizing the luckiness normalized maximum likelihood (LNML) code-length. The causal relationship between each variable pair is determined by selecting the shortest code-length of the causal model, and we introduce the notion of $Δ$-pseudo-collinearity to identify dependencies induced by latent confounders. Based on these ideas, we develop a greedy algorithm, termed Pseudo-Collinearity Guided Causal Discovery (PCG-CD). Experiments on synthetic and real-world datasets demonstrate that the proposed method accurately recovers directed causal relationships and effectively detects latent confounders.","short_abstract":"Causal discovery with nonlinear mechanisms and latent confounders remains challenging. Existing methods often rely on either linear assumptions or causal sufficiency, limiting their applicability. We propose an MDL-based causal discovery framework that explicitly accounts for latent confounders while allowing flexible...","url_abs":"https://arxiv.org/abs/2607.04133","url_pdf":"https://arxiv.org/pdf/2607.04133v1","authors":"[\"Zhongyi Que\",\"Shin Matsushima\",\"Kenji Yamanishi\"]","published":"2026-07-05T06:19:48Z","proceeding":"cs.LG","tasks":"[\"cs.LG\"]","methods":"[]","has_code":false}