Characterizing Fairness Over the Set of Good Models Under Selective Labels

Amanda Coston, Ashesh Rambachan, Alexandra Chouldechova
Proceedings of the 38th International Conference on Machine Learning, PMLR 139:2144-2155, 2021.

Abstract

Algorithmic risk assessments are used to inform decisions in a wide variety of high-stakes settings. Often multiple predictive models deliver similar overall performance but differ markedly in their predictions for individual cases, an empirical phenomenon known as the “Rashomon Effect.” These models may have different properties over various groups, and therefore have different predictive fairness properties. We develop a framework for characterizing predictive fairness properties over the set of models that deliver similar overall performance, or “the set of good models.” Our framework addresses the empirically relevant challenge of selectively labelled data in the setting where the selection decision and outcome are unconfounded given the observed data features. Our framework can be used to 1) audit for predictive bias; or 2) replace an existing model with one that has better fairness properties. We illustrate these use cases on a recidivism prediction task and a real-world credit-scoring task.

Cite this Paper

BibTeX
@InProceedings{pmlr-v139-coston21a, title = {Characterizing Fairness Over the Set of Good Models Under Selective Labels}, author = {Coston, Amanda and Rambachan, Ashesh and Chouldechova, Alexandra}, booktitle = {Proceedings of the 38th International Conference on Machine Learning}, pages = {2144--2155}, year = {2021}, editor = {Meila, Marina and Zhang, Tong}, volume = {139}, series = {Proceedings of Machine Learning Research}, month = {18--24 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v139/coston21a/coston21a.pdf}, url = {https://proceedings.mlr.press/v139/coston21a.html}, abstract = {Algorithmic risk assessments are used to inform decisions in a wide variety of high-stakes settings. Often multiple predictive models deliver similar overall performance but differ markedly in their predictions for individual cases, an empirical phenomenon known as the “Rashomon Effect.” These models may have different properties over various groups, and therefore have different predictive fairness properties. We develop a framework for characterizing predictive fairness properties over the set of models that deliver similar overall performance, or “the set of good models.” Our framework addresses the empirically relevant challenge of selectively labelled data in the setting where the selection decision and outcome are unconfounded given the observed data features. Our framework can be used to 1) audit for predictive bias; or 2) replace an existing model with one that has better fairness properties. We illustrate these use cases on a recidivism prediction task and a real-world credit-scoring task.} }
Endnote
%0 Conference Paper %T Characterizing Fairness Over the Set of Good Models Under Selective Labels %A Amanda Coston %A Ashesh Rambachan %A Alexandra Chouldechova %B Proceedings of the 38th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2021 %E Marina Meila %E Tong Zhang %F pmlr-v139-coston21a %I PMLR %P 2144--2155 %U https://proceedings.mlr.press/v139/coston21a.html %V 139 %X Algorithmic risk assessments are used to inform decisions in a wide variety of high-stakes settings. Often multiple predictive models deliver similar overall performance but differ markedly in their predictions for individual cases, an empirical phenomenon known as the “Rashomon Effect.” These models may have different properties over various groups, and therefore have different predictive fairness properties. We develop a framework for characterizing predictive fairness properties over the set of models that deliver similar overall performance, or “the set of good models.” Our framework addresses the empirically relevant challenge of selectively labelled data in the setting where the selection decision and outcome are unconfounded given the observed data features. Our framework can be used to 1) audit for predictive bias; or 2) replace an existing model with one that has better fairness properties. We illustrate these use cases on a recidivism prediction task and a real-world credit-scoring task.
APA
Coston, A., Rambachan, A. & Chouldechova, A.. (2021). Characterizing Fairness Over the Set of Good Models Under Selective Labels. Proceedings of the 38th International Conference on Machine Learning, in Proceedings of Machine Learning Research 139:2144-2155 Available from https://proceedings.mlr.press/v139/coston21a.html.

Related Material