This paper introduces a new framework for understanding and mitigating fairness issues in sequential decision-making systems. It categorizes uncertainty into model, feedback, and prediction types, highlighting how these can disproportionately affect under-represented groups. The research demonstrates that by accounting for unequal uncertainty and selective feedback, it's possible to reduce outcome variance for disadvantaged populations while still meeting institutional goals. AI
Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →
IMPACT Provides a framework for auditing and governing fairness risks in sequential AI decision systems, particularly where uncertainty is unevenly distributed.
RANK_REASON This is a research paper published on arXiv detailing a new framework for fairness in sequential decision-making.
