Research map / Learning theory and algorithms / Fairness
Predict Responsibly: Improving Fairness and Accuracy by Learning to Defer
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 |
|---|---|---|---|
| Adapt deferral decisions to the decision-maker's predictions and weaknesses instead of using fixed rejection | Learning to defer · COMPAS, high-accuracy DM | accuracy-fairness tradeoff | Helped |
| Adapt deferral decisions to the decision-maker's bias | Learning to defer · COMPAS, highly-biased DM | accuracy-fairness tradeoff | Helped |
| Choose cases for deferral based on the decision-maker's subgroup-specific reliability | Learning to defer · COMPAS and Heritage Health, inconsistent DM | accuracy | Helped |
| Use an adaptive model that takes X into account when deciding whether to defer, rather than post-hoc thresholding based only on model scores | Learning to defer · Inconsistent DM scenario | accuracy | Helped |
| Add the decision-maker's extra information to the decision pipeline through adaptive deferral | Learning to defer · COMPAS and Heritage Health, inconsistent DM | accuracy: first 10% -> first ~40% | Helped |
| Defer more on subgroups where the decision-maker is reliable and less where unreliable | Learning to defer · COMPAS and Heritage Health, inconsistent DM | accuracy | Helped |
| Adapt deferral to the decision-maker's reliability while evaluating minimum subgroup accuracy | Learning to defer · COMPAS and Heritage Health, inconsistent DM | minimum subgroup accuracy | Helped |
| Use rejection learning without modeling the decision-maker | Rejection learning · Inconsistent DM scenario | accuracy | No clear effect |
| Use a constant-loss oracle decision-maker during training (ablation) | Learning to defer · Oracle-DM comparison | accuracy-fairness tradeoff | No clear effect |
Closest papers by meaning
- A Unifying Post-Processing Framework for Multi-Objective Learn-to-Defer Problems (NeurIPS 2024 · similarity 0.86)
- Fair Secretaries with Unfair Predictions (NeurIPS 2024 · similarity 0.85)
- Fair Off-Policy Learning from Observational Data (ICML 2024 · similarity 0.85)
- Adapting Static Fairness to Sequential Decision-Making: Bias Mitigation Strategies towards Equal Long-term Benefit Rate (ICML 2024 · similarity 0.85)
- Understanding Fairness and Prediction Error through Subspace Decomposition and Influence Analysis (NeurIPS 2025 · similarity 0.85)
- Fairness without Harm: An Influence-Guided Active Sampling Approach (NeurIPS 2024 · 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.