Fairness-Aware Estimation of Graphical Models

2024/08/30 by Zhou, Zhuoping, Tarzanagh, Davoud Ataee, Hou, Bojian +2 · 3 citations
#FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML)

paper · doi:10.48550/arxiv.2408.17396

Abstract

This paper examines the issue of fairness in the estimation of graphical models (GMs), particularly Gaussian, Covariance, and Ising models. These models play a vital role in understanding complex relationships in high-dimensional data. However, standard GMs can result in biased outcomes, especially when the underlying data involves sensitive characteristics or protected groups. To address this, we introduce a comprehensive framework designed to reduce bias in the estimation of GMs related to protected attributes. Our approach involves the integration of the pairwise graph disparity error and a tailored loss function into a nonsmooth multi-objective optimization problem, striving to achieve fairness across different sensitive groups while maintaining the effectiveness of the GMs. Experimental evaluations on synthetic and real-world datasets demonstrate that our framework effectively mitigates bias without undermining GMs' performance.

Cited by

• Learning Conditional Independence Differential Graphs From Time-Dependent Data
• Estimating Fair Graphs from Graph-Stationary Data
• Fair Minimum Labeling: Efficient Temporal Network Activations for Reachability and Equity

Related

• Fair GLASSO: Estimating Fair Graphical Models with Unbiased Statistical Behavior
• The KL-Divergence between a Graph Model and its Fair I-Projection as a Fairness Regularizer
• Fairness Constraints in High-Dimensional Generalized Linear Models
• Finding Representative Group Fairness Metrics Using Correlation Estimations
• Fairness criteria through the lens of directed acyclic graphical models