Statistical and Structural Approaches to Algorithmic Fairness
Antonio Ferrara
cs.LG, cs.AI, stat.ML
Submitted: 2026-06-24
Comments: Doctoral thesis
Code: https://github.com/alanturin-g/SAFT
Project page: https://christophm.github.io/interpretable-ml-book
License: http://creativecommons.org/licenses/by/4.0/
The gist: Modern machine learning systems have outgrown their origins as isolated predictive constructs, evolving into complex socio-technical architectures that actively mediate human opportunity.
Terminology
Abstract
Modern machine learning systems have outgrown their origins as isolated predictive constructs, evolving into complex socio-technical architectures that actively mediate human opportunity. As algorithms increasingly determine access to economic and social opportunities, it has become widely recognized that these systems are deeply embedded with the structural inequalities and prejudices of their environments. The field of algorithmic fairness emerged in response to the growing recognition that models optimized for predictive accuracy can systematically disadvantage marginalized groups. Early mitigation strategies, however, rested on fragile simplifications that limited their effectiveness in complex socio-technical environments. This thesis identifies and addresses two fundamental limitations of contemporary fairness paradigms: the reliance on deterministic point estimates for auditing and the treatment of individuals as isolated entities devoid of structural context.
Sources
- Confidence Intervals for Testing Disparate Impact in Fair Learning
- Auditing and Achieving Intersectional Fairness in Classification Problems
- hyppo: A Multivariate Hypothesis Testing Python Package
- Statistical Inference for Fairness Auditing
- Modeling Tabular data using Conditional GAN
- True to the Model or True to the Data?
- Do We Really Need Deep Learning Models for Time Series Forecasting?
- Explaining black box decisions by Shapley cohort refinement
- Explanation Shift: Detecting distribution shifts on tabular data via the explanation space
- Towards the Right Kind of Fairness in AI
- A Survey of Shortest-Path Algorithms
- Expected Maximin Fairness in Max-Cut and other Combinatorial Optimization Problems
- The Effect of People Recommenders on Echo Chambers and Polarization
- Exposure Inequality in People Recommender Systems: The Long-Term Effects
- node2vec: Scalable Feature Learning for Networks
- General Geospatial Inference with a Population Dynamics Foundation Model
- Aggregation of pairwise comparisons with reduction of biases
- Best-Worst Scaling More Reliable than Rating Scales: A Case Study on Sentiment Intensity Annotation
- PRD: Peer Rank and Discussion Improve Large Language Model based Evaluations
- IMDB-WIKI-SbS: An Evaluation Dataset for Crowdsourced Pairwise Comparisons
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks