NeuFair: Neural Network Fairness Repair with Dropout

TL;DR

NeuFair introduces a post-processing dropout-based method to mitigate bias in pre-trained neural networks, effectively improving fairness with minimal impact on accuracy by selectively dropping neurons during inference.

Contribution

This work presents NeuFair, a novel randomized dropout algorithm that enhances fairness in pre-trained DNNs without retraining or modifying original data or algorithms.

Findings

Achieves up to 69% fairness improvement

Maintains model utility with minimal performance loss

Outperforms existing bias mitigation methods

Abstract

This paper investigates neuron dropout as a post-processing bias mitigation for deep neural networks (DNNs). Neural-driven software solutions are increasingly applied in socially critical domains with significant fairness implications. While neural networks are exceptionally good at finding statistical patterns from data, they may encode and amplify existing biases from the historical data. Existing bias mitigation algorithms often require modifying the input dataset or the learning algorithms. We posit that the prevalent dropout methods that prevent over-fitting during training by randomly dropping neurons may be an effective and less intrusive approach to improve the fairness of pre-trained DNNs. However, finding the ideal set of neurons to drop is a combinatorial problem. We propose NeuFair, a family of post-processing randomized algorithms that mitigate unfairness in pre-trained DNNs via dropouts during inference after training. Our randomized search is guided by an objective to minimize discrimination while maintaining the model's utility. We show that our design of randomized algorithms is effective and efficient in improving fairness (up to 69%) with minimal or no model performance degradation. We provide intuitive explanations of these phenomena and carefully examine the influence of various hyperparameters of search algorithms on the results. Finally, we empirically and conceptually compare NeuFair to different state-of-the-art bias mitigators.