Flavio Calmon, “Information-Theoretic Methods for Trustworthy Machine Learning”
Please join us on Tuesday, September 26, 2023 at 12:00pm in Clark Hall Room 110 and on ZOOM for the
CIS & MINDS Seminar Series:
Guest: Flavio Calmon
Assistant Professor, Electrical Engineering
Harvard University
Topic: “Information-Theoretic Methods for Trustworthy Machine Learning”
Join Zoom Meeting: https://wse.zoom.us/j/95386212146
Passcode: cis&minds
If you would like to meet with Flavio Calmon, please sign up at this link: TBA
Flavio Calmon
Assistant Professor, Electrical Engineering
Harvard University
“Information-Theoretic Methods for Trustworthy Machine Learning”
Abstract: This talk overviews recent information-theoretic results in three aspects of trustworthy machine learning: group fairness, predictive multiplicity, and differential privacy. First, we briefly overview a post-processing technique, “FairProjection,” designed to ensure group fairness in prediction and classification. We then present converse results based on Blackwell’s “comparison of experiments” theorem that captures the limits of group-fairness assurance in classification. These results show that existing techniques (including FairProjection) can approach the optimal Pareto frontier between accuracy and group fairness in specific settings.
Second, we review the concept of predictive multiplicity in machine learning. Predictive multiplicity arises when different classifiers achieve similar average performance for a specific learning task, yet produce conflicting predictions for individual samples. We discuss a metric called “Rashomon Capacity” for quantifying predictive multiplicity in multi-class classification. We also present recent findings on the multiplicity cost of differentially private training methods and fairness interventions in machine learning.
Finally, we briefly highlight recent results at the interface of information theory and differential privacy. This includes new methods for privacy accounting and the design of privacy mechanisms.
This talk is based on work published at NeurIPS’22, ACM FAccT’23, and ICML’23.
Biography: Flavio P. Calmon is an Assistant Professor of Electrical Engineering at the Harvard John A. Paulson School of Engineering and Applied Sciences. Before joining Harvard, he was the inaugural Data Science for Social Good Post-Doctoral Fellow at IBM Research in Yorktown Heights, New York. He received his Ph.D. in Electrical Engineering and Computer Science at MIT. His research develops information-theoretic tools for responsible and reliable machine learning. Prof. Calmon has received the NSF CAREER award, faculty awards from Google, IBM, and Amazon, the NSF-Amazon Fairness in AI award, the Harvard Data Science Initiative Bias2 award, and the Harvard Dean of Undergraduate Studies Commendation for “Extraordinary Teaching during Extraordinary Times.” He also received the inaugural Título de Honra ao Mérito (Honor to the Merit Title) given to alumni from the Universidade de Brasília (Brazil).