Research map / Graph learning / Graph algorithms and clustering
DAGs with No Curl: An Efficient DAG Structure Learning Approach
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| initialize W and p randomly instead of using the Step 1 estimate (ablation) | DAG-NoCurl · ER6-Gaussian d=30 | SHD: 21.44 ± 1.56 -> 84.88 ± 2.65 | Hurt |
| omit Step 1 and start with random p (ablation) | DAG-NoCurl · ER6-Gaussian d=30 | SHD: 17.37 ± 1.18 -> 130.37 ± 1.66 | Hurt |
| omit Step 2 and rely on thresholding until acyclic (ablation) | DAG-NoCurl · ER6-Gaussian d=30 | SHD: 21.44 ± 1.56 -> 37.36 ± 1.34; 17.37 ± 1.18 -> 29.88 ± 1.55 | Hurt |
| use the alternative objective to optimize W instead of the fixed-p objective (ablation) | DAG-NoCurl · ER6-Gaussian d=30 | SHD: 21.44 ± 1.56 -> 32.82 ± 1.07; 17.37 ± 1.18 -> 26.08 ± 1.07 | Hurt |
| add joint optimization of W and p after Step 2 (ablation) | DAG-NoCurl · ER6-Gaussian d=30 | SHD: 21.44 ± 1.56 -> 21.44 ± 1.56; 17.37 ± 1.18 -> 17.81 ± 1.29 | No clear effect |
| add joint optimization of W and p after Step 2 (ablation) | DAG-NoCurl · ER6-Gaussian d=30 | Time (Sec): 3.34 ± 0.21 -> 5.32 ± 0.35; 7.68 ± 0.39 -> 10.38 ± 0.23 | Hurt |
Closest papers by meaning
- CoLiDE: Concomitant Linear DAG Estimation (ICLR 2024 · similarity 0.89)
- Analytic DAG Constraints for Differentiable DAG Learning (ICLR 2025 · similarity 0.89)
- Markov Equivalence and Consistency in Differentiable Structure Learning (NeurIPS 2024 · similarity 0.89)
- Boosting Causal Structure Learning: An Asymmetric Exponential Modulation Gaussian-Based Adaptive Sample Reweighting Framework (AAAI 2025 · similarity 0.88)
- Constraint-Free Structure Learning with Smooth Acyclic Orientations (ICLR 2024 · similarity 0.88)
- ProDAG: Projected Variational Inference for Directed Acyclic Graphs (NeurIPS 2025 · similarity 0.88)
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.