Research map / Probabilistic and scientific modeling / Causal discovery
Distinguishing Cause and Effect in Bivariate Structural Causal Models: A Systematic Investigation
Christoph Käding, Jakob Runge
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 |
|---|---|---|---|
| add the large bSCMC collection to the weighted average | SLOPPY · Weighted average accuracy | accuracy: 52.14 % -> 85.72 % | Improved |
| add the large bSCMC collection to the weighted average | RECI · Weighted average accuracy | accuracy: 58.62 % -> 83.47 % | Improved |
| add the large bSCMC collection to the weighted average | GPI · Weighted average accuracy | accuracy: 63.89 % -> 72.26 % | Improved |
| add the large bSCMC collection to the weighted average | CGNN · Weighted average accuracy | accuracy: 64.44 % -> 76.00 % | Improved |
| weight by number of data sets instead of plain averaging | LiNGAM · All collections incl. bSCMC | accuracy: 45.94 % -> 29.43 % | Hurt |
| weight by number of data sets instead of plain averaging | RCC · All collections incl. bSCMC | accuracy: 51.81 % -> 34.30 % | Hurt |
| use newer larger TCEP version (99 pairs instead of 68) | LiNGAM · TCEP | accuracy: ~62% -> ~45% | Hurt |
| use weighted TCEP (wTCEP) instead of TCEP | RECI and Jarfo · TCEP vs wTCEP | accuracy | Improved |
| use weighted TCEP (wTCEP) instead of TCEP | PNL · TCEP vs wTCEP | accuracy | Hurt |
| include bSCMC in the unweighted average | GPI · Avg accuracy | accuracy: 71.99 % -> 72.06 % | No clear effect |
| include bSCMC in the unweighted average | CGNN · Avg accuracy | accuracy: 70.74 % -> 71.65 % | Improved |
| vary functional dependence type (mul function with bimodal normal cause) | CGNN vs GPI · bSCMC | accuracy | Mixed |
Closest papers by meaning
- Detecting and Measuring Confounding Using Causal Mechanism Shifts (NeurIPS 2024 · similarity 0.90)
- Distinguishing Cause from Effect with Causal Velocity Models (ICML 2025 · similarity 0.89)
- Identifying Structural Biases from Causal Mechanism Shifts (NeurIPS 2026 · similarity 0.88)
- Bivariate Causal Discovery using Bayesian Model Selection (ICML 2024 · similarity 0.88)
- Data Fusion for Partial Identification of Causal Effects (NeurIPS 2025 · similarity 0.88)
- Do Contemporary Causal Inference Models Capture Real-World Heterogeneity? Findings from a Large-Scale Benchmark (ICLR 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.
