Adaptive Fairness Improvement Based on Causality Analysis

TL;DR

This paper introduces an adaptive approach that uses causality analysis to select the most effective fairness improvement method for neural networks, balancing fairness and accuracy efficiently.

Contribution

It proposes a novel causality-based adaptive method for fairness enhancement that outperforms existing static approaches in effectiveness and efficiency.

Findings

Effectively identifies the best fairness method for given models

Maintains high accuracy while improving fairness

Operates with an average overhead of 5 minutes

Abstract

Given a discriminating neural network, the problem of fairness improvement is to systematically reduce discrimination without significantly scarifies its performance (i.e., accuracy). Multiple categories of fairness improving methods have been proposed for neural networks, including pre-processing, in-processing and post-processing. Our empirical study however shows that these methods are not always effective (e.g., they may improve fairness by paying the price of huge accuracy drop) or even not helpful (e.g., they may even worsen both fairness and accuracy). In this work, we propose an approach which adaptively chooses the fairness improving method based on causality analysis. That is, we choose the method based on how the neurons and attributes responsible for unfairness are distributed among the input attributes and the hidden neurons. Our experimental evaluation shows that our approach is effective (i.e., always identify the best fairness improving method) and efficient (i.e., with an average time overhead of 5 minutes).