Can You Fake It Until You Make It? - Impacts of Differentially Private Synthetic Data on Downstream Classification Fairness

This course explores the impacts of differentially private synthetic data on downstream classification fairness. The learning outcomes include understanding the concept of differentially private synthetic data and its effects on classification fairness. The course teaches the skills of analyzing experimental results and drawing conclusions based on the findings. The teaching method involves presenting a proof, discussing an experimental pipeline, and showcasing results. The intended audience for this course is individuals interested in data privacy, synthetic data generation, and fairness in classification algorithms.