Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Kabra, Anmol, Karzand, Mina, Lechner, Tosca, Srebro, Nathan, Wang, Serena
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2410.06290
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866916428582486016
author Kabra, Anmol
Karzand, Mina
Lechner, Tosca
Srebro, Nathan
Wang, Serena
author_facet Kabra, Anmol
Karzand, Mina
Lechner, Tosca
Srebro, Nathan
Wang, Serena
contents We present a framework for designing scores to summarize performance metrics. Our design has two multi-criteria objectives: (1) improving on scores should improve all performance metrics, and (2) achieving pareto-optimal scores should achieve pareto-optimal metrics. We formulate our design to minimize the dimensionality of scores while satisfying the objectives. We give algorithms to design scores, which are provably minimal under mild assumptions on the structure of performance metrics. This framework draws motivation from real-world practices in hospital rating systems, where misaligned scores and performance metrics lead to unintended consequences.
format Preprint
id arxiv_https___arxiv_org_abs_2410_06290
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Score Design for Multi-Criteria Incentivization
Kabra, Anmol
Karzand, Mina
Lechner, Tosca
Srebro, Nathan
Wang, Serena
Computers and Society
Computational Geometry
Machine Learning
We present a framework for designing scores to summarize performance metrics. Our design has two multi-criteria objectives: (1) improving on scores should improve all performance metrics, and (2) achieving pareto-optimal scores should achieve pareto-optimal metrics. We formulate our design to minimize the dimensionality of scores while satisfying the objectives. We give algorithms to design scores, which are provably minimal under mild assumptions on the structure of performance metrics. This framework draws motivation from real-world practices in hospital rating systems, where misaligned scores and performance metrics lead to unintended consequences.
title Score Design for Multi-Criteria Incentivization
topic Computers and Society
Computational Geometry
Machine Learning
url https://arxiv.org/abs/2410.06290