Mathematical Proofs for Fair AI Bias Analysis

25 Mar 2025

This appendix provides mathematical proofs for bias in DP-based fair learning and additional results on image dataset fairness using ResNet-18 models.

How to Reduce Majority Bias in AI Models

25 Mar 2025

This study explores bias in DP-based fair learning and proposes SA-DRO to reduce majority attribute bias while maintaining accuracy in AI models.

Achieving Fair AI Without Sacrificing Accuracy

24 Mar 2025

SA-DRO reduces bias in DP-based fair learning models, improving accuracy in federated learning while balancing fairness across sensitive attributes.

How to Test for AI Fairness

24 Mar 2025

Explore how AI fairness is tested using COMPAS, Adult, and CelebA datasets, with DP-based learning methods and neural networks to measure bias in classification

The Limits of Demographic Parity in AI Models

24 Mar 2025

Discover the hidden biases in DP-based fair learning and how distributionally robust optimization (DRO) can improve AI fairness and reduce bias in ML

How to Measure Fairness in AI Models

24 Mar 2025

Explore fairness criteria in AI and how dependence measures help detect bias in supervised learning models, ensuring ethical and responsible machine learning.

What to Do When ‘Fair’ AI Delivers Unfair Results

24 Mar 2025

Discover how SA-DRO reduces bias in fair machine learning models by addressing demographic parity flaws.

Research Suggests AI Models Can Deliver More Accurate Diagnoses Without Discrimination

31 Dec 2024

A new study introduces the concept of positive-sum fairness in AI models for medical imaging, suggesting that larger performance gaps aren't always harmful.

How AI Models Could Detect Lung Conditions Fairly

31 Dec 2024

Study shows how positive-sum fairness improves lung lesion detection without harming subgroup performance.