Research map / Learning theory and algorithms / Fairness
Demystifying Algorithmic Fairness in an Uncertain World
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
A research statement/talk abstract surveying the author's three works on algorithmic fairness under data (aleatoric) and model (epistemic) uncertainty, including uncertainty-aware fairness and coverage-based fairness (EOC).
Problem
Fair ML methods implicitly assume certain and reliable environments; fairness measures vary highly across training runs, and bias arises from both data and model uncertainty.
Approach
Use predictive uncertainty decomposition (aleatoric vs. epistemic) to understand and improve algorithmic fairness, and propose Equal Opportunity of Coverage (EOC) requiring equal prediction-interval coverage across groups.
Limitations
- Text is a short research statement with no experimental details
- Surveyed works are only summarized, not evaluated here
Closest papers by meaning
- Equal Opportunity of Coverage in Fair Regression (NeurIPS 2023 · similarity 0.90)
- Aleatoric and Epistemic Discrimination: Fundamental Limits of Fairness Interventions (NeurIPS 2023 · similarity 0.89)
- Reconciling Predictive and Statistical Parity: A Causal Approach (AAAI 2024 · similarity 0.88)
- Achievable Fairness on Your Data With Utility Guarantees (NeurIPS 2024 · similarity 0.88)
- The Fragility of Fairness: Causal Sensitivity Analysis for Fair Machine Learning (NeurIPS 2024 · similarity 0.88)
- Geometry of Relaxed Fair Regression: A Unified Framework for Aware and Unaware Settings (NeurIPS 2026 · 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.