Fairness with Exponential Weights

Document detail

ID

oai:arXiv.org:2411.04295

Topic

Computer Science - Machine Learnin...

Author

Pasteris, Stephen Hicks, Chris Mavroudis, Vasilios

Year

2024

listing date

2/19/2025

Keywords

algorithm

Metrics

Abstract

Motivated by the need to remove discrimination in certain applications, we develop a meta-algorithm that can convert any efficient implementation of an instance of Hedge (or equivalently, an algorithm for discrete bayesian inference) into an efficient algorithm for the equivalent contextual bandit problem which guarantees exact statistical parity on every trial.

Relative to any comparator with statistical parity, the resulting algorithm has the same asymptotic regret bound as running the corresponding instance of Exp4 for each protected characteristic independently.

Given that our Hedge instance admits non-stationarity we can handle a varying distribution with which to enforce statistical parity with respect to, which is useful when the true population is unknown and needs to be estimated from the data received so far.

Via online-to-batch conversion we can handle the equivalent batch classification problem with exact statistical parity, giving us results that we believe are novel and important in their own right.

Pasteris, Stephen,Hicks, Chris,Mavroudis, Vasilios, 2024, Fairness with Exponential Weights