Research map / Probabilistic and scientific modeling / Causal discovery
A Simultaneous Discover-Identify Approach to Causal Inference in Linear Models
Chi Zhang, Bryant Chen, Judea Pearl
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use LCDI instead of FCI to learn arrowheads | LCDI · Random DAGs, avg degree (4,4.5], n=6 to 11 | % DAGs with more arrowheads: 0 -> 45.0 | Improved |
| Increase number of nodes from 6 to 11 with LCDI vs FCI | LCDI · Random DAGs, avg degree (2.75,3.25] | % DAGs with more arrowheads: 5.5 -> 25.5 | Improved |
| Increase number of nodes from 6 to 11 with LCDI vs FCI | LCDI · Random DAGs, avg degree (1.5,2] | % DAGs with more arrowheads: 1.0 -> 2.0 | Mixed |
| Increase average degree from (1.5,2] to (4,4.5] at n=11 | LCDI · Random DAGs n=11 | % DAGs with more arrowheads: 2.0 -> 45.0 | Improved |
| LCDI vs FCI on arrowhead gain for small degree | LCDI · Random DAGs avg degree (1.5,2] | % more arrowheads when LCDI learns more: 17.4 | Improved |
| Increase number of nodes at large degree | LCDI · Random DAGs avg degree (4,4.5] | % more arrowheads when LCDI wins: 8.7 -> 6.8 | Hurt |
| LCDI vs FCI on extraneous edges | LCDI · Random DAGs n=6 to 11 | number of extraneous edges | Improved |
| Increase n and d for FCI extraneous edges | LCDI · Random DAGs | number of extraneous edges (FCI) | Hurt |
Closest papers by meaning
- Causal learning with the invariance principle (NeurIPS 2026 · similarity 0.89)
- Efficient Identification in Linear Structural Causal Models with Instrumental Cutsets (NeurIPS 2019 · similarity 0.89)
- Strong and Weak Identifiability of Optimization-based Causal Discovery in Non-linear Additive Noise Models (ICML 2025 · similarity 0.88)
- Differentiable Causal Discovery for Latent Hierarchical Causal Models (ICLR 2025 · similarity 0.88)
- Identification of Causal Structure with Latent Variables Based on Higher Order Cumulants (AAAI 2024 · similarity 0.88)
- Identifying Causal Effects Under Functional Dependencies (NeurIPS 2024 · 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.
