FETA: Fairness Enforced Verifying, Training, and Predicting Algorithms for Neural Networks
This work addresses fairness in algorithmic decision-making for applications affecting people's quality of life, offering a novel approach to guarantee fairness beyond existing methods.
The paper tackles the challenge of ensuring individual fairness in neural network predictions by developing a counterexample-guided post-processing technique that provably enforces fairness constraints at prediction time across the entire input domain, and an in-processing method that incorporates fairness counterexamples during training, resulting in models with significantly higher fairness and accuracy on real-world datasets.
Algorithmic decision making driven by neural networks has become very prominent in applications that directly affect people's quality of life. In this paper, we study the problem of verifying, training, and guaranteeing individual fairness of neural network models. A popular approach for enforcing fairness is to translate a fairness notion into constraints over the parameters of the model. However, such a translation does not always guarantee fair predictions of the trained neural network model. To address this challenge, we develop a counterexample-guided post-processing technique to provably enforce fairness constraints at prediction time. Contrary to prior work that enforces fairness only on points around test or train data, we are able to enforce and guarantee fairness on all points in the input domain. Additionally, we propose an in-processing technique to use fairness as an inductive bias by iteratively incorporating fairness counterexamples in the learning process. We have implemented these techniques in a tool called FETA. Empirical evaluation on real-world datasets indicates that FETA is not only able to guarantee fairness on-the-fly at prediction time but also is able to train accurate models exhibiting a much higher degree of individual fairness.