Individual Risk: CS 333 / STAT 333  

Co-Taught with Prof. Joe Blitzstein

 

Course Description: Risk assessment tools are increasingly deployed in high-stakes settings: What is the probability that the tumor will metastasize?  What is the chance that this individual will commit a violent crime in the next two years?  What is the probability that the student will graduate within 4 years?  But what is the probability of a non-repeatable event?  What is the mathematical meaning of "individual risk" and what should we require of a risk assessment algorithm? This reading course will explore different notions of risk, based on different notions of probability, and will connect this literature to notions of regret and indistinguishability from computer science.

 

Fairness and Privacy: Perspectives of Law and Probability: CS 126

Intertwined with course of same name taught by Prof. Martha Minow, Harvard Law School

 

Course Description: Students will learn to analyze and mitigate privacy loss, overfitting, and unfairness in data analysis. Principal techniques will come from cryptography, differential privacy, and the newly emerging areas of adaptive data analysis and algorithmic fairness.

Note: Offered jointly by HLS and SEAS, with interwoven tracks emphasizing, respectively, law and computer science, the tracks will meet jointly and separately every week.