Fairness and Bias in AI Systems

The field of artificial intelligence is moving towards a greater emphasis on fairness and bias detection. Researchers are investigating how AI systems can perpetuate and amplify existing social biases, and are developing new methods to detect and mitigate these biases. One of the key areas of focus is on the development of more inclusive and diverse datasets, as well as the creation of new fairness metrics and evaluation frameworks. Additionally, there is a growing interest in the use of crowdsourced human judgment and perception-driven approaches to detect bias in AI systems. Noteworthy papers in this area include:

  • Perception-Driven Bias Detection in Machine Learning via Crowdsourced Visual Judgment, which introduces a novel framework for bias detection that leverages crowdsourced human judgment.
  • Evaluating Fairness and Mitigating Bias in Machine Learning: A Novel Technique using Tensor Data and Bayesian Regression, which proposes a new technique for evaluating fairness in machine learning for image classification tasks.
  • DETONATE: A Benchmark for Text-to-Image Alignment and Kernelized Direct Preference Optimization, which introduces a large-scale benchmark for text-to-image alignment and kernelized direct preference optimization.

Sources

When Algorithms Play Favorites: Lookism in the Generation and Perception of Faces

Perception-Driven Bias Detection in Machine Learning via Crowdsourced Visual Judgment

Gender Fairness of Machine Learning Algorithms for Pain Detection

Evaluating Fairness and Mitigating Bias in Machine Learning: A Novel Technique using Tensor Data and Bayesian Regression

Hidden Bias in the Machine: Stereotypes in Text-to-Image Models

Public Acceptance of Cybernetic Avatars in the service sector: Evidence from a Large-Scale Survey in Dubai

One Size Fits None: Rethinking Fairness in Medical AI

DETONATE: A Benchmark for Text-to-Image Alignment and Kernelized Direct Preference Optimization

Built with on top of