AI Paper Atlas

Research map / Probabilistic and scientific modeling / Causal discovery

Boosting Causal Additive Models

Maximilian Kertel, Nadja Klein

JMLR 2025 · Probabilistic and scientific modeling › Causal discovery · approach: Constraint-based structure learning with latents

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
increase sample size N from 10 to 200boosting causal-order score · Synthetic additive SEs, p=5transposition distance: 3.29 -> 0.03Improved
increase sample size N from 10 to 200 with non-additive SEsboosting causal-order score · Synthetic non-additive SEs, p=5transposition distance: 3.26 -> 0.23Improved
DAGBoost versus DAGSNOTEARSDAGBoost · Synthetic additive SF, p=100SHD: 176.96 -> 30.08Improved
DAGBoost versus CAM on non-additive SF graphsDAGBoost · Synthetic non-additive SF, p=100SHD: 77.63 -> 67.47Improved
DAGBoost versus CAM on additive ER graphsDAGBoost · Synthetic additive ER, p=100SHD: 12.07 -> 14.90Hurt
DAGBoost versus CAM on additive SF graphsDAGBoost · Synthetic additive SF, p=100SHD: 30.18 -> 30.08No clear effect
DAGBoost versus CAM, precisionDAGBoost · Synthetic non-additive SF, p=100precision: 0.676 -> 0.930Improved
DAGBoost versus CAM, recallDAGBoost · Synthetic additive SF, p=100recall: 0.817 -> 0.712Hurt
increase boosting step size from 0.3 to 0.9 (ablation)DAGBoost · Synthetic ER additive, p=100runtime (s): 58.47 -> 20.17Improved
increase boosting step size from 0.3 to 0.9 (ablation)DAGBoost · Synthetic ER additive, p=100SHD: 14.90 -> 13.17No clear effect
vary penalty lambda from 0.001 to 0.1 (ablation)DAGBoost · Synthetic ER additive, p=100SHD: 15.58 -> 14.90No clear effect
increase penalty lambda from 0.001 to 0.1 (ablation)DAGBoost · Synthetic ER additive, p=100runtime (s): 43.54 -> 204.87Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy