Research map / Probabilistic and scientific modeling / Causal discovery
Causal normalizing flows: from theory to practice
Adrián Javaloy, Pablo Sanchez Martin, Isabel Valera
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 causal NF (abductive, L=1) instead of CAREFL with do-operator | CAREFL · Triangle NLIN | counterfactual RMSE: 0.17_{0.03} -> 0.13_{0.02} | Improved |
| use causal NF instead of CAREFL | CAREFL · Large backdoor NLIN | ATE RMSE: 0.05_{0.01} -> 0.02_{0.00} | Improved |
| use causal NF instead of CAREFL | CAREFL · Large backdoor NLIN | counterfactual RMSE: 0.08_{0.01} -> 0.01_{0.00} | Improved |
| use causal NF instead of CAREFL | CAREFL · Simpson SYMPROD | ATE RMSE: 0.10_{0.02} -> 0.07_{0.01} | Improved |
| use causal NF instead of CAREFL | CAREFL · Large backdoor NLIN | sampling time (microseconds): 8.25_{1.29} -> 3.05_{0.66} | Improved |
| use causal NF instead of VACA | VACA · Triangle NLIN | KL divergence: 7.71_{0.60} -> 0.00_{0.00} | Improved |
| use causal NF instead of VACA | VACA · Large backdoor NLIN | training time (microseconds): 164.92_{11.10} -> 0.52_{0.10} | Improved |
| use causal NF instead of CAREFL | CAREFL · Large backdoor NLIN | KL divergence: 1.51_{0.05} -> 1.51_{0.04} | No clear effect |
| use abductive instead of generative network design (ablation) | causal NF · 4-chain SCM | KL divergence, ATE, counterfactual RMSE | Improved |
| embed causal graph in abductive architecture (ablation) | causal NF · 4-chain SCM | ATE RMSE | Improved |
| generative models with causal graph and fewer layers than diameter (L<3) (ablation) | causal NF · 4-chain SCM | KL divergence | Hurt |
| add Jacobian regularization (ablation) | abductive causal NF L=1 with graph · 4-chain SCM | causal consistency | No clear effect |
Closest papers by meaning
- Learning Structural Causal Models from Ordering: Identifiable Flow Models (AAAI 2025 · similarity 0.91)
- Causally Consistent Normalizing Flow (AAAI 2025 · similarity 0.90)
- Marginal Causal Flows for Validation and Inference (NeurIPS 2024 · similarity 0.87)
- CausalTab: Pretraining Across Causal Environments for Tabular Causal Discovery (NeurIPS 2026 · similarity 0.86)
- A Practical Approach to Causal Inference over Time (AAAI 2025 · similarity 0.86)
- Learning Linear Causal Representations from General Environments: Identifiability and Intrinsic Ambiguity (NeurIPS 2024 · similarity 0.86)
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.
