Research map / Probabilistic and scientific modeling / Causal discovery
Kolmogorov-Arnold Causal Generative Models
Alejandro Almodóvar, Mar Elizo, Patricia A. Apellániz, Santiago Zazo, Juan Parras
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 KaCGM mixed-type variants (KAN_mix) instead of KAN on discrete data | KaCGM · Synthetic graphs, mixed type | MMD observational p-value: 0.042 -> 0.690 | Improved |
| use ANM instead of KaCGM KAN on continuous data | KaCGM (KAN) · Synthetic graphs, continuous | MMD observational (x1e-3): 16.49 -> 8.86 | Hurt |
| use KaCGM (KAAM) instead of DBCM | DBCM · Synthetic graphs, continuous | MMD observational (x1e-3): 31.50 -> 16.76 | Improved |
| use more flexible CFlow instead of KaCGM when noise is nonadditive | KaCGM (KAN) · Sachs nonadditive | MMD observational (x1e-3): 5.27 -> 1.42 | Hurt |
| use KaCGM in additive Sachs vs nonadditive Sachs | KaCGM (KAN) · Sachs | MMD observational (x1e-3): 1.36 -> 5.27 | Hurt |
| reduce the number of training samples from 1000 to 10 | KaCGM · Synthetic graphs | all metrics | Hurt |
| apply pruning to learned KANs (ablation) | KaCGM · Cardiovascular dataset | observational fit | No clear effect |
| apply symbolic substitution after pruning (ablation) | KaCGM · Cardiovascular dataset | observational fit | Hurt |
| increase nonadditivity alpha of the noise (ablation) | KaCGM · Sensitivity DGP, alpha 0 to 1 | observational MMD and independence | Hurt |
Closest papers by meaning
- CausalTab: Pretraining Across Causal Environments for Tabular Causal Discovery (NeurIPS 2026 · similarity 0.88)
- A Fixed-Point Approach for Causal Generative Modeling (ICML 2024 · similarity 0.88)
- Learning Mixtures of Unknown Causal Interventions (NeurIPS 2024 · similarity 0.88)
- Conditional Generative Models are Sufficient to Sample from Any Causal Effect Estimand (NeurIPS 2024 · similarity 0.88)
- Controllable Generative Sandbox for Causal Inference (NeurIPS 2026 · similarity 0.88)
- Learning Structural Causal Models from Ordering: Identifiable Flow Models (AAAI 2025 · similarity 0.87)
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.
