Research map / Probabilistic and scientific modeling / Causal discovery
Equality of Opportunity in Classification: A Causal Approach
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 |
|---|---|---|---|
| constrain disparate ER to half (f_er) instead of unconstrained | standard fairness classifier · Synthetic standard prediction model | accuracy: 90.4% -> 90.4% | No clear effect |
| constrain disparate ER to half (f_er) instead of unconstrained | standard fairness classifier · Synthetic standard prediction model | counterfactual direct ER: 0.381 -> 0.620 | Hurt |
| constrain counterfactual ERs to half (f_ctf) instead of only the disparate ER | standard fairness classifier · Synthetic standard prediction model | counterfactual direct ER: 0.620 -> 0.191 | Helped |
| constrain counterfactual ERs (f_ctf) instead of disparate ER (f_er) | standard fairness classifier · Synthetic standard prediction model | accuracy: 90.4% -> 90.1% | No clear effect |
| constrain disparate ER to |ER|<=0.01 (f_er) instead of unconstrained | COMPAS classifier · COMPAS | accuracy: 74.6% -> 73.7% | Hurt |
| constrain disparate ER to |ER|<=0.01 (f_er) instead of unconstrained | COMPAS classifier · COMPAS | disparate ER (false positive rate): −0.077 -> -0.005 | Helped |
| remove prior convictions W from features and constrain direct ER (Alg. 3, f_ctf-) | COMPAS classifier · COMPAS | counterfactual direct ER: 0.015 -> -0.001 | Helped |
| remove prior convictions W and add direct-ER constraint (f_ctf-) vs f_er | COMPAS classifier · COMPAS | accuracy: 73.7% -> 72.1% | Hurt |
Closest papers by meaning
- Fairness-Accuracy Trade-Offs: A Causal Perspective (AAAI 2025 · similarity 0.87)
- Reconciling Predictive and Statistical Parity: A Causal Approach (AAAI 2024 · similarity 0.87)
- FairPFN: A Tabular Foundation Model for Causal Fairness (ICML 2025 · similarity 0.86)
- Causal Bias Detection in Generative Artifical Intelligence (NeurIPS 2026 · similarity 0.86)
- Local Causal Discovery for Structural Evidence of Direct Discrimination (AAAI 2025 · similarity 0.86)
- Fairness on Principal Stratum: A New Perspective on Counterfactual Fairness (ICML 2025 · 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.