Research map / Probabilistic and scientific modeling / Causal discovery
Distributionally Robust Causal Abstractions
Yorgos Felekis, Theodoros Damoulas, Paris Giampouras
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 best-tuned robustness radii instead of theoretical radii (ablation) | DiRoCA · EBM (full corruption) | average abstraction error: 297.2 -> 268.4 | Improved |
| use best-tuned robustness radii instead of theoretical radii (ablation) | DiRoCA · cMNIST (full corruption) | average abstraction error: 215.0 -> 48.2 | Improved |
| use best-tuned robustness radii instead of theoretical radii (ablation) | DiRoCA · LiLUCAS (omega misspecification) | average abstraction error: 459.1 -> 304.1 | Improved |
| use best-tuned robustness radii instead of theoretical radii (ablation) | DiRoCA · nLUCAS (omega misspecification) | average abstraction error: 18.2 -> 9.1 | Improved |
| use best-tuned robustness radii instead of theoretical radii (ablation) | DiRoCA · cMNIST rotation 90 deg | average abstraction error: 106.4 -> 30.7 | Improved |
| add distributional robustness (DiRoCA theoretical radii vs non-robust Grad) | DiRoCA · LiLUCAS (omega misspecification) | average abstraction error: 305.5 -> 459.1 | Hurt |
| add distributional robustness (DiRoCA best radii vs non-robust Grad) | DiRoCA · cMNIST (full corruption) | average abstraction error: 22200.7 -> 48.2 | Improved |
| add distributional robustness (DiRoCA best radii vs non-robust Grad) | DiRoCA · EBM (full corruption) | average abstraction error: 320.2 -> 268.4 | Improved |
| add distributional robustness (DiRoCA best radii vs non-robust Grad) | DiRoCA · cMNIST rotation 30 deg | average abstraction error: 1780.9 -> 28.0 | Improved |
| add distributional robustness at low contamination | DiRoCA · SLC / LiLUCAS (alpha near 0) | average abstraction error | Hurt |
Closest papers by meaning
- Designing Ambiguity Sets for Distributionally Robust Optimization Using Structural Causal Optimal Transport (AAAI 2025 · similarity 0.88)
- Causal Abstractions, Categorically Unified (NeurIPS 2026 · similarity 0.87)
- Causal Abstraction Learning based on the Semantic Embedding Principle (ICML 2025 · similarity 0.87)
- Learning Structural Causal Models from Ordering: Identifiable Flow Models (AAAI 2025 · similarity 0.86)
- Robust agents learn causal world models (ICLR 2024 · similarity 0.86)
- Fine-Grained Causal Dynamics Learning with Quantization for Improving Robustness in Reinforcement Learning (ICML 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.
