Canonical_sets

Description:

 AI systems can create, propagate, support, and automate bias in decision-making processes. To mitigate biased decisions, we both need to understand the origin of the bias and define what it means for an algorithm to make fair decisions. By Locating Unfairness through Canonical Inverse Design (LUCID), we generate a canonical set that shows the desired inputs for a model given a preferred output. The canonical set reveals the model's internal logic and exposes potential unethical biases by repeatedly interrogating the decision-making process.
 
LUCID-GAN extends on LUCID by generating canonical inputs via a conditional generative model instead of gradient-based inverse design. LUCID-GAN generates canonical inputs conditional on the predictions of the model under fairness evaluation. LUCID-GAN has several benefits, including that it applies to non-differentiable models, ensures that a canonical set consists of realistic inputs, and allows us to assess indirect discrimination and explicitly check for intersectional unfairness.

More Information:

• Originates from the Edge AI research challenge (FAIR 2019–2023).

• Responsible Research Lead: Vincent Ginis

• Authors:Andres Algaba, Carmen Mazijn, Vincent Ginis

• Programming Language: Python

• Link to the repository

• Link to the website

• Categories: Supervised Learning, Images

Publications:

• Mazijn, C., Prunkl, C., Algaba, A., Danckaert, J., & Ginis, V. (2023). LUCID: Exposing algorithmic bias through inverse design. In Proceedings of the Thirty-Seventh AAAI Conference on Artificial Intelligence. https://doi.org/10.1609/aaai.v37i12.26683 

• Algaba, A., Mazijn, C., Prunkl, C., Danckaert, J., Ginis, V. (2023). LUCID–GAN: Conditional Generative Models to Locate Unfairness. In Explainable Artificial Intelligence. xAI 2023. https://doi.org/10.1007/978-3-031-44070-0_18