Constructing Fair Latent Space for Intersection of Fairness and Explainability

2024-12-23

AI Summary

Addressing the challenge of simultaneously achieving fairness and interpretability in machine learning models, this paper proposes a latent-space fairification method. It lightweightly integrates a disentanglement module into a pre-trained generative model to explicitly separate label-relevant from sensitive-attribute-relevant representations and re-distribute the sensitive subspace—thereby enforcing algorithmic fairness at the latent-space level. Crucially, the method enables generation of faithful, fair, and verifiable counterfactual explanations without fine-tuning or retraining the entire generative model. Experiments demonstrate substantial improvements in fairness metrics—including demographic parity and equalized odds—while providing auditable, transparent decision rationales. To our knowledge, this is the first work to achieve plug-and-play joint optimization of fairness and interpretability on pre-trained generative models.

Technology Category

• Machine Learning: Transparent, Interpretable, Explainable ML
• Computer Vision: Interpretability, Explainability, and Transparency
• Humans and AI: Explainable AI (XAI) for Human Understanding

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Large pretrained models with web data

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.

Problem

• Fairness
• Interpretability
• Trust enhancement

Innovation

• Fairness Enhancement
• Interpretability Improvement
• Efficient Training