Explore the critical intersection of privacy and fairness in machine learning algorithms in this thought-provoking conference talk from Strange Loop. Dive into the importance of preserving privacy in data science and its role in creating more ethical and fair models. Examine research on fair-and-private machine learning algorithms and privacy-preserving models, demonstrating how prioritizing user privacy in machine learning workflows can lead to better overall models and support more ethical product design. Investigate topics such as predictive policing, privacy as privilege and power, learning fair representations, and privacy by design. Gain insights into why privacy matters in the age of big data and increased data protection regulations, and how it contributes to creating more equitable algorithmic outcomes.
Overview
Syllabus
Intro
Data Science: The Solution to Social Ills
Predictive Policing
Privacy as Privilege
Privacy as Power
Privacy and Fairness
Learning Fair Representations
Private and Fair Representations
Private and Fair Machine Learning
Privacy By Design
Taught by
Strange Loop Conference
