AI Paper Atlas

Research map / Probabilistic and scientific modeling / Causal discovery

Better Simulations for Validating Causal Discovery with the DAG-Adaptation of the Onion Method

Bryan Andrews, Erich Kummerfeld

JMLR 2026 · Probabilistic and scientific modeling › Causal discovery · approach: Differentiable and score-based DAG learning · 6 citations (OpenAlex)

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
simulate with DaO instead of ZARXZARX simulation · ER-DAGs |V|=20 (R2-sortability)R2 rank correlation: -0.859 -> -0.264Improved
simulate with DaO instead of TetradTetrad simulation · ER-DAGs |V|=100 (R2-sortability)R2 rank correlation: -0.557 -> -0.386Improved
simulate with DaO instead of ZARXZARX simulation · SFo-DAGs |V|=100 (R2-sortability)R2 rank correlation: -0.497 -> -0.084Improved
simulate with DaO instead of ZARXZARX simulation · standardized beta coefficients, 100 variablesabsolute beta coefficientsMixed
simulate with DaO instead of ZARXZARX simulation · error term standard deviationsstandard deviationImproved
lower threshold on beta coefficients from 0.3 to 0.1 (ablation)DAGMA · DaO and ZARX standardized modelsedge recovery thresholdImproved
standardize data before DAGMA (sDAGMA) (ablation)DAGMA · ZARX ER-DAG simulations, 20 variablesstructure recoveryMixed
move from ZARX to DaO simulationPC and GES · ER-DAGs 20 variablesstructure recoveryImproved
change DAG structure from ER to SFoDaO simulation · R2-sortabilityR2 rank correlation: -0.264 -> 0.033Improved

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