Constructing Fair Latent Space for Intersection of Fairness and Explainability

TL;DR

This paper introduces a module that creates a fair latent space in generative models, improving fairness and explainability simultaneously, with minimal additional training and computational costs.

Contribution

It proposes a novel method to construct a fair latent space by disentangling labels and sensitive attributes, enhancing both fairness and explainability in generative models.

Findings

Effective fairness improvement demonstrated by various metrics

Provides faithful explanations for biased decisions

Reduces training time and costs by only training the module

Abstract

As the use of machine learning models has increased, numerous studies have aimed to enhance fairness. However, research on the intersection of fairness and explainability remains insufficient, leading to potential issues in gaining the trust of actual users. Here, we propose a novel module that constructs a fair latent space, enabling faithful explanation while ensuring fairness. The fair latent space is constructed by disentangling and redistributing labels and sensitive attributes, allowing the generation of counterfactual explanations for each type of information. Our module is attached to a pretrained generative model, transforming its biased latent space into a fair latent space. Additionally, since only the module needs to be trained, there are advantages in terms of time and cost savings, without the need to train the entire generative model. We validate the fair latent space with various fairness metrics and demonstrate that our approach can effectively provide explanations for biased decisions and assurances of fairness.