Introducing Positive-Sum Fairness: A New Way to Balance Performance and Equity in Medical AI | HackerNoon
Briefly

from Hackernoon1 month ago
Positive-sum fairness is a fairness evaluation framework where the goal is to find solutions that increase the overall benefit for all parties together while trying to ensure no one is worse off and ideally, everyone is better off.
Hackernoonhttps://hackernoon.com/introducing-positive-sum-fairness-a-new-way-to-balance-performance-and-equity-in-medical-ai
When looking at changes in model performance and disparities between protected subgroups, there are several explanations for a gap in performance between the most and least advantaged subgroups.
Hackernoonhttps://hackernoon.com/introducing-positive-sum-fairness-a-new-way-to-balance-performance-and-equity-in-medical-ai
Unlike other fairness definitions which aim to minimize the disparity between subgroups or maximize the worst performance among subgroups, positive-sum fairness tries to avoid gains to a group which come at the expense of another group while maintaining the overall performance.
Hackernoonhttps://hackernoon.com/introducing-positive-sum-fairness-a-new-way-to-balance-performance-and-equity-in-medical-ai
The most advantaged group's performance improved while others' stayed the same, or all subgroups' performance improved but one of them increased more than others, would not be considered harmful.
Hackernoonhttps://hackernoon.com/introducing-positive-sum-fairness-a-new-way-to-balance-performance-and-equity-in-medical-ai
Read at Hackernoon
#fairness #model-performance #disparities #protected-subgroups #evaluation-framework
[
|
]