AI Paper Atlas

Research map / Probabilistic and scientific modeling / Causal discovery

Structural Agnostic Modeling: Adversarial Learning of Causal Graphs

Diviyan Kalainathan, Olivier Goudet, Isabelle Guyon, David Lopez-Paz, Michèle Sebag

JMLR 2022 · 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
add adversarial loss instead of MSE loss (ablation)SAM · Synthetic 20-variable graphs (Global)AUPR: 0.41 -> 0.57Improved
add non-linear mechanisms instead of linear ones (ablation)SAM · Synthetic 20-variable graphs (Global)AUPR: 0.34 -> 0.57Improved
add both non-linear mechanisms and adversarial loss (ablation)SAM · Synthetic 20-variable graphs (Global)AUPR: 0.30 -> 0.57Improved
add adversarial loss instead of MSE loss (ablation)SAM · Synthetic 100-variable graphs (Global)AUPR: 0.25 -> 0.59Improved
add non-linear mechanisms instead of linear ones (ablation)SAM · Synthetic 100-variable graphs (Global)AUPR: 0.39 -> 0.59Improved
add both non-linear mechanisms and adversarial loss (ablation)SAM · Synthetic 100-variable graphs (Global)AUPR: 0.17 -> 0.59Improved
add adversarial loss instead of MSE loss (ablation)SAM · Synthetic 20-variable graphs (Global)SHD: 43.9 -> 35.9Improved
add non-linear mechanisms and adversarial loss on linear data (ablation)SAM · Synthetic 20-variable graphs (Linear)AUPR: 0.30 -> 0.32No clear effect
add the acyclicity (DAG) constraint (ablation)SAM · Synthetic 20-variable GP Mix graphs, 500 samplesAUPRImproved
average over more runs, keep majority edges (ablation)SAM · Synthetic 20-variable GP graphs, 500 samplesAUPRImproved
increase graph density from 0.1 to 0.2SAM · Synthetic 20-variable GP graphs, 500 samplesAUPRImproved
increase graph density above 0.2SAM · Synthetic 20-variable GP graphs, 500 samplesAUPRHurt

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