Unfair Utilities and First Steps Towards Improving Them
This work addresses fairness issues in machine learning policies, particularly in domains like criminal justice, but it is incremental as it builds on existing fairness frameworks.
The paper tackles the problem of fairness in policy optimization by proposing to focus on modifying utility functions to satisfy a new criterion called value of information fairness, rather than constraining predictors, and demonstrates that this approach provides better answers than existing fairness notions in thought experiments and the COMPAS data.
Many fairness criteria constrain the policy or choice of predictors, which can have unwanted consequences, in particular, when optimizing the policy under such constraints. Here, we advocate to instead focus on the utility function the policy is optimizing for. We define value of information fairness and propose to not use utility functions that violate this criterion. This principle suggests to modify these utility functions such that they satisfy value of information fairness. We describe how this can be done and discuss consequences for the corresponding optimal policies. We apply our framework to thought experiments and the COMPAS data. Focussing on the utility function provides better answers than existing fairness notions: We are not aware of any intuitively fair policy that is disallowed by value of information fairness, and when we find that value of information fairness recommends an intuitively unfair policy, no existing fairness notion finds an intuitively fair policy.