mlsanityData to 7 Oct 2026

Fairness Invariants: A Relational Approach to Explaining and Mitigating Fairness Bugs

arXiv 2608.26209 ↗First posted 26 Aug 2026cs.SE cs.AI cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Data-driven software systems are increasingly deployed in high-stakes socio-economic domains, from criminal justice to financial lending. However, these systems often exhibit individual discrimination---unjustified disparities in which a program yields different outcomes for similar individuals who differ only in their protected attributes (e.g., race, gender, age). While existing research has focused on detecting and quantifying these bugs, there remains a critical lack of principled mechanisms to explain and localize individual fairness bugs. Current explanation techniques are largely designed for single-input decisions rather than the relational nature of discrimination, which inherently involves a comparison between an original and a counterfactual pair. We present REMI, a framework for the automated localization, explanation, and mitigation of individual discrimination. Inspired by loop-invariant synthesis in formal methods, we treat counterfactual fairness as a relational invariant discovery problem. We introduce a bidirectional relational explanation framework that learns over paired examples (x, x') to identify regions of the input space where fairness is violated. Unlike traditional one-way implication pairs used in invariant inference, our approach enforces bidirectional constraints: requiring identical outcomes for both original and counterfactual samples. REMI utilizes three data-alignment techniques to infer interpretable rule-based models that act as "fairness invariants." These rules serve as guardrails to selectively block or relabel unfair predictions without requiring model retraining. Our evaluation on symbolic and neural network programs demonstrates that REMI localizes ground-truth fairness bugs in over 83% of cases, significantly outperforming state-of-the-art baselines and reducing discriminatory decisions in black-box models by up to 70%.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 36 of its 69 references that are in the map. First posted on arXiv 26 Aug 2026.

Builds on

36 of 69 references.

References · 36

  1. 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
  2. 2016XGBoost: A Scalable Tree Boosting Systemcited by 5,120
  3. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,040
  4. 2016Equality of Opportunity in Supervised Learningcited by 2,000
  5. 2017Counterfactual Fairnesscited by 802
  6. 2018Mitigating Unwanted Biases with Adversarial Learningcited by 663
  7. 2019Explaining Machine Learning Classifiers through Diverse Counterfactual Explanationscited by 612
  8. 2018A Reductions Approach to Fair Classificationcited by 585
  9. 2018AI Fairness 360: An Extensible Toolkit for Detecting, Understanding, and Mitigating Unwanted Algorithmic Biascited by 319
  10. 2017Fairness Testing: Testing Software for Discriminationcited by 123
  11. 2021Bias in Machine Learning Software: Why? How? What to do?cited by 92
  12. 2018Automated Directed Fairness Testingcited by 77
24 more
  1. 2020Verifying Individual Fairness in Machine Learning Modelscited by 49
  2. 2022Causality-based Neural Network Repaircited by 44
  3. 2019Perfectly Parallel Fairness Certification of Neural Networkscited by 42
  4. 2021NeuronFair: Interpretable White-Box Fairness Testing through Biased Neuron Identificationcited by 42
  5. 2023Exploring and Interacting with the Set of Good Sparse Generalized Additive Modelscited by 30
  6. 2022Fairness-aware Configuration of Machine Learning Librariescited by 28
  7. 2023Latent Imitator: Generating Natural Individual Discriminatory Instances for Black-Box Fairness Testingcited by 23
  8. 2022Fairify: Fairness Verification of Neural Networkscited by 22
  9. 2018Automated Test Generation to Detect Individual Discrimination in AI Modelscited by 21
  10. 2022FairNeuron: Improving Deep Neural Network Fairness with Adversary Games on Selective Neuronscited by 20
  11. 2019Software Engineering for Fairness: A Case Study with Hyperparameter Optimizationcited by 18
  12. 2022Adaptive Fairness Improvement Based on Causality Analysiscited by 17
  13. 2020VerifyMed -- A blockchain platform for transparent trust in virtualized healthcare: Proof-of-conceptcited by 14
  14. 2022Fast Interpretable Greedy-Tree Sumscited by 12
  15. 2023Fix Fairness, Don't Ruin Accuracy: Performance Aware Fairness Repair using AutoMLcited by 9
  16. 2024NeuFair: Neural Network Fairness Repair with Dropoutcited by 9
  17. 2023Certifying the Fairness of KNN in the Presence of Dataset Biascited by 7
  18. 2024FairLay-ML: Intuitive Debugging of Fairness in Data-Driven Social-Critical Softwarecited by 4
  19. 2024MAFT: Efficient Model-Agnostic Fairness Testing for Deep Neural Networks via Zero-Order Gradient Searchcited by 4
  20. 2025Attention Pruning: Automated Fairness Repair of Language Models via Surrogate Simulated Annealingcited by 4
  21. 2024Predicting Fairness of ML Software Configurationscited by 1
  22. 2025Uncovering Discrimination Clusters: Quantifying and Explaining Systematic Fairness Violationscited by 1
  23. 2026On the Robustness of Fairness Practices: A Causal Framework for Systematic Evaluationcited by 1
  24. 2026Analyzing Fairness of Neural Network Prediction via Counterfactual Dataset Generationcited by 1

Built on by

Not cited yet.