AI Paper Atlas

Research map / Graph learning / Graph algorithms and clustering

Revisiting Differentiable Structure Learning: Inconsistency of L1 Penalty and Beyond

Kaifeng Jin, Ignavier Ng, Kun Zhang, Biwei Huang

AAAI 2026 · Graph learning › Graph algorithms and clustering · approach: Continuous DAG structure learning · 2 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
add the estimated moral graph to restrict the search space (ablation)CALM (hard constraint) · 50-node ER1 (infinite samples, standardized)SHD of CPDAG: 16.7 -> 5.5Improved
add the estimated moral graph to restrict the search space (ablation)soft-constraint ℓ0 likelihood · 50-node ER1 (infinite samples, standardized)SHD of CPDAG: 33.8 -> 7.6Improved
use hard instead of soft DAG constraint (ablation)ℓ0 likelihood with moral graph · 50-node ER1 (infinite samples, standardized)SHD of CPDAG: 7.6 -> 5.5Improved
use hard instead of soft DAG constraint (no moral graph) (ablation)ℓ0 likelihood without moral graph · 50-node ER1 (infinite samples, standardized)SHD of CPDAG: 33.8 -> 16.7Improved
standardize the data before CALM (ablation)CALM · 50-node ER1 (infinite samples)SHD of CPDAG: 9.9 -> 5.5No clear effect
replace ℓ1-penalized GOLEM-NV with CALM (ℓ0 Gumbel-Softmax, hard DAG, moral graph)GOLEM-NV-ℓ1 · 50-node ER1 (n=1000, standardized)SHD of CPDAG: 60.0 -> 12.1Improved
replace ℓ1-penalized GOLEM-NV with CALMGOLEM-NV-ℓ1 · 50-node ER4 (n=1000, standardized)SHD of CPDAG: 211.6 -> 168.8Improved
replace ℓ1-penalized GOLEM-NV with CALMGOLEM-NV-ℓ1 · Sachs real dataSHD of CPDAG: 13 -> 12Improved
use CALM instead of discrete PC/FGES on sparse graphs50-node ER1 (n=1000) · 50-node ER1 (n=1000, standardized)SHD of CPDAG: CALM 12.1 vs PC 11.0, FGES 8.4Hurt
use CALM instead of PC/FGES on dense graphs50-node ER4 (n=1000) · 50-node ER4 (n=1000, standardized)SHD of CPDAG: PC 200.5 / FGES 425.6 -> CALM 168.8Improved
use ℓ1 instead of ℓ0 penalty (minimum-ℓ1 DAG among covariance-consistent DAGs) (ablation)linear Gaussian structure learning · 8-node ER1 population covariancenumber of edges / SHD of CPDAG: edges 8.0 -> 22.74; SHD 0 -> 19.97Hurt

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