The Sharpe predictor for fairness in machine learning
This work addresses fairness issues in machine learning that can discriminate against minority groups, offering a novel approach to integrate fairness meaningfully, though it appears incremental as it builds on existing SMOO concepts from finance.
The authors tackled the problem of fairness in machine learning by introducing a new paradigm based on Stochastic Multi-Objective Optimization (SMOO), which allows simultaneous optimization of accuracy and fairness metrics, leading to the discovery of the complete trade-off landscape and the introduction of the Sharpe predictor that provides the highest accuracy-to-unfairness ratio.
In machine learning (ML) applications, unfair predictions may discriminate against a minority group. Most existing approaches for fair machine learning (FML) treat fairness as a constraint or a penalization term in the optimization of a ML model, which does not lead to the discovery of the complete landscape of the trade-offs among learning accuracy and fairness metrics, and does not integrate fairness in a meaningful way. Recently, we have introduced a new paradigm for FML based on Stochastic Multi-Objective Optimization (SMOO), where accuracy and fairness metrics stand as conflicting objectives to be optimized simultaneously. The entire trade-offs range is defined as the Pareto front of the SMOO problem, which can then be efficiently computed using stochastic-gradient type algorithms. SMOO also allows defining and computing new meaningful predictors for FML, a novel one being the Sharpe predictor that we introduce and explore in this paper, and which gives the highest ratio of accuracy-to-unfairness. Inspired from SMOO in finance, the Sharpe predictor for FML provides the highest prediction return (accuracy) per unit of prediction risk (unfairness).