Project 4: Machine Learning Fairness

Project Description

Term: Spring 2023

• Team #1
• Projec title: Machine Learning Fairness Exploration(LFR, C-LR, C-SVM)
• Team members
  • Cao, Shengqi [email protected]
  • Li, Haoyang [email protected]
  • Li, Yuanxi [email protected]
  • Liu, Kejun [email protected]
  • Ng, Brendan [email protected]
  • Zhang, Zixun [email protected]
• Project summary: In this project, our group studied 3 machine learning fairness methods on 2 papers, the Learning Fair Representations (LFR) methods from Zemel et.al, and Constrained Support Vector Machines (C-SVM) and Constrained Logistic Regression (CLR) from Zafar et.al. By using the COMPAS dataset and implementing the methods in the paper, we try to find out the relation between the two-year recid with some other attributes using machine learning algorithms that take fairness into account. Finally, we evaluated those models on four metrics: accuracy, calibration, equality of odds, and parity.

Contribution statement: [default] All team members contributed equally in all stages of this project. All team members approve our work presented in this GitHub repository including this contributions statement. Zixun, Haoyang and Yuanxi contributed to the LFR model (A1). Brendan contributed to the C-LR implementation, while Shengqi and Kejun contributed to the C-SVM model (A2). Lastly, Haoyang is the presenter of our project, and Shengqi wrote down this Readme file.

Following suggestions by RICH FITZJOHN (@richfitz). This folder is orgarnized as follows.

proj/
├── lib/
├── data/
├── doc/
├── figs/
└── output/

Please see each subfolder for a README file.