Fairness and privacy, Removing context encourages unfairness
Han Zhao @hanzhao_ml How to ensure fairness (statistical parity) and privacy (DP) simultaneously? What are the costs of privacy and fairness upon accuracy? Excited to share our #ICML2024 work answering the two questions above!
paper: https://arxiv.org/pdf/2405.04034
code: https://github.com/rxian/fair-regression
Replying to @hanzhao_ml
Han, Link to abstracts to give context. If I could, I would outlaw use of PDF on the Internet, because it strips the information needed for your type of algorithm, then copies propagate. LLMs are like that too. Richard Collins, The Internet Foundation
https://arxiv.org/abs/2405.04034
