FairBED: Incorporating Fairness into Bayesian Experimental Design
stat.ML, cs.LG
Submitted: 2026-06-22
Updated: 2026-10-18
License: http://creativecommons.org/licenses/by/4.0/
The gist: Frameworks for ensuring fairness in machine learning typically focus on learning fair models from existing data.
Terminology
Abstract
Frameworks for ensuring fairness in machine learning typically focus on learning fair models from existing data. But this endeavor is often undermined by biases already present in that data. We therefore look to modify the data acquisition process itself to help gather fairer data that is inherently more suitable for training fair predictors. To this end, we introduce FairBED, which provides novel formulations for quantifying the fairness of datasets themselves based on the idea that fair datasets should be uninformative about sensitive attributes. We then use this to construct practical fairness-aware Bayesian experimental design (BED) objectives that maximize expected information gain about the target quantity of interest while minimizing expected information gain about sensitive attributes. We further derive a theoretical link between FairBED and demographic parity, and show empirically that models trained on data gathered using FairBED provide improved fairness-accuracy trade-offs compared to randomly acquired data and conventional BED.
Sources
- Active Sampling for Min-Max Fairness
- Fairness in Clustering with Multiple Sensitive Attributes
- Performance Comparisons of Reinforcement Learning Algorithms for Sequential Experimental Design
- AI Fairness 360: An Extensible Toolkit for Detecting, Understanding, and Mitigating Unwanted Algorithmic Bias
- Making Better Use of Unlabelled Data in Bayesian Active Learning
- Fair prediction with disparate impact: A study of bias in recidivism prediction instruments
- Unprocessing Seven Years of Algorithmic Fairness
- Review of Mathematical frameworks for Fairness in Machine Learning
- Fairness Through Awareness
- Unbiased MLMC stochastic gradient-based optimization of Bayesian experimental designs
- Bayesian Active Learning for Classification and Preference Learning
- ALINE: Joint Amortization for Bayesian Inference and Active Data Acquisition
- Bayesian Experimental Design via Contrastive Diffusions
- Nesting Particle Filters for Experimental Design in Dynamical Systems
- Policy-Based Bayesian Experimental Design for Non-Differentiable Implicit Models
- Falcon: Fair Active Learning using Multi-armed Bandits
- From Selection to Generation: A Survey of LLM-based Active Learning
Related papers
- Behavior of prediction performance metrics with rare events
- Optimal Estimation of Generic Dynamics by Path-Dependent Neural Jump ODEs
- A Posterior-Dynamics Framework for Imaging Inverse Problems with Pretrained Diffusion Priors
- Scalable Posterior Uncertainty for Flexible Density-Based Clustering
- A Transport Geometry of Benign Overfitting
- Efficient quadratic entropy with distance sketches