Research map / Probabilistic and scientific modeling / Causal discovery
VACA: Designing Variational Graph Autoencoders for Causal Queries
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 |
|---|---|---|---|
| Use VACA instead of CAREFL | Causal-query estimation · Collider LIN | observational MMD: 9.27±1.49 -> 1.50±0.67 | Helped |
| Use VACA instead of CAREFL | Causal-query estimation · Collider NLIN | interventional MMD: 4.69±0.38 -> 0.97±0.23 | Helped |
| Use VACA instead of CAREFL | Causal-query estimation · Loan | observational MMD: 22.10±1.64 -> 2.22±0.25 | Helped |
| Use VACA instead of CAREFL | Causal-query estimation · Loan | interventional MMD: 27.38±4.07 -> 6.87±0.66 | Helped |
| Use VACA instead of CAREFL | Causal-query estimation · Adult | observational MMD: 31.31±1.58 -> 4.51±0.45 | Helped |
| Use VACA instead of CAREFL | Causal-query estimation · Adult | counterfactual MSE: 1.23±0.17 -> 5.33±0.27 | Hurt |
| Exclude the sensitive attribute from inputs | SVM on German Credit · German Credit | counterfactual unfairness: 14.01 ± 2.26 -> 13.27 ± 2.28 | No clear effect |
| Restrict inputs to non-descendants of the sensitive attribute | SVM on German Credit · German Credit | counterfactual unfairness: 14.01 ± 2.26 -> 0.14 ± 0.02 | Helped |
| Use VACA latent variables excluding the sensitive-attribute latent | SVM on German Credit · German Credit | f1: 71.67 -> 70.79 ± 5.15 | No clear effect |
| Use VACA latent variables excluding the sensitive-attribute latent | SVM on German Credit · German Credit | counterfactual unfairness: 14.01 ± 2.26 -> 0.51 ± 0.19 | Helped |
Closest papers by meaning
- DeepITE: Designing Variational Graph Autoencoders for Intervention Target Estimation (NeurIPS 2024 · similarity 0.88)
- Interventional Fairness on Partially Known Causal Graphs: A Constrained Optimization Approach (ICLR 2024 · similarity 0.87)
- A Local Method for Satisfying Interventional Fairness with Partially Known Causal Graphs (NeurIPS 2024 · similarity 0.87)
- Causal Inference with Conditional Front-Door Adjustment and Identifiable Variational Autoencoder (ICLR 2024 · similarity 0.86)
- FairPFN: A Tabular Foundation Model for Causal Fairness (ICML 2025 · similarity 0.85)
- A Recipe for Causal Graph Regression: Confounding Effects Revisited (ICML 2025 · similarity 0.85)
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.