Research map / Probabilistic and scientific modeling / Causal discovery
Counterfactual Fairness with Partially Known Causal Graph
Aoqi Zuo, Susan Wei, Tongliang Liu, Bo Han, Kun Zhang, Mingming Gong
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 FairRelax (add possible descendants) instead of Fair | counterfactual fair prediction · Synthetic, 10 nodes | RMSE: 1.137$\pm$0.824 -> 1.031$\pm$0.751 | Improved |
| use FairRelax instead of Fair | counterfactual fair prediction · Synthetic, 30 nodes | RMSE: 1.024$\pm$0.908 -> 0.797$\pm$0.489 | Improved |
| use FairRelax instead of Fair | counterfactual fair prediction · Synthetic, 40 nodes | RMSE: 0.800$\pm$0.480 -> 0.755$\pm$0.441 | Improved |
| use FairRelax instead of Fair | counterfactual fair prediction · Synthetic, 10 nodes | unfairness: 0.000$\pm$0.000 -> 0.023$\pm$0.123 | Hurt |
| use FairRelax instead of Unaware | linear regression predictor · Synthetic, 10 nodes | unfairness: 0.200$\pm$0.322 -> 0.023$\pm$0.123 | Improved |
| use FairRelax instead of Full | linear regression predictor · Synthetic, 10 nodes | unfairness: 0.288$\pm$0.363 -> 0.023$\pm$0.123 | Improved |
| use FairRelax instead of Full | linear regression predictor · Synthetic, 10 nodes | RMSE: 0.621$\pm$0.251 -> 1.031$\pm$0.751 | Hurt |
| use Fair instead of Full | linear regression predictor · UCI Student | unfairness: 0.761$\pm$0.228 -> 0.000$\pm$0.000 | Improved |
| use Fair instead of Full | linear regression predictor · UCI Student | RMSE: 3.49$\pm$0.292 -> 3.509$\pm$0.356 | Hurt |
| use Fair instead of Unaware | linear regression predictor · UCI Student | unfairness: 0.250$\pm$0.085 -> 0.000$\pm$0.000 | Improved |
| drop sensitive attribute (Unaware vs Full) | linear regression predictor · UCI Student | unfairness: 0.761$\pm$0.228 -> 0.250$\pm$0.085 | Improved |
| drop sensitive attribute (Unaware vs Full) | linear regression predictor · UCI Student | RMSE: 3.49$\pm$0.292 -> 3.482$\pm$0.297 | No clear effect |
Closest papers by meaning
- Interventional Fairness on Partially Known Causal Graphs: A Constrained Optimization Approach (ICLR 2024 · similarity 0.94)
- Towards counterfactual fairness through auxiliary variables (ICLR 2025 · similarity 0.91)
- A Local Method for Satisfying Interventional Fairness with Partially Known Causal Graphs (NeurIPS 2024 · similarity 0.90)
- FairPFN: A Tabular Foundation Model for Causal Fairness (ICML 2025 · similarity 0.90)
- Causal Bias Detection in Generative Artifical Intelligence (NeurIPS 2026 · similarity 0.89)
- Counterfactual Fairness by Combining Factual and Counterfactual Predictions (NeurIPS 2024 · 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.
