Connections between Online Learning and Differential Privacy

Abstract: Online learning is a branch of machine learning that deals with algorithms that process data incrementally. Often these algorithms are analyzed in a probability-free game-theoretic framework so that their guarantees hold not just for iid data but even for data generated adversarially. This is in contrast to standard statistical learning theory which is formulated in terms of a batch of data drawn iid from some distribution. Recently, there have been efforts to modify such batch algorithm to make them differentially private. This is done so that we can preserve the privacy of individuals whose data is part of the training set. On the surface, online learning and differentially private learning have little to do with each other. However, I will try to convince you that online and differentially private learning are fundamentally connected. The connection rests on a key property of good algorithms in both area: stability to change of one example in the training dataset.

(Talk is based on joint work with Jacob Abernethy, Young Hun Jung, Baekjin Kim, Chansoo Lee, and Audra McMillan.)

Speaker Bio:  Ambuj Tewari is an associate professor in the Department of Statistics and the Department of EECS (by courtesy) at the University of Michigan, Ann Arbor. He is also affiliated with the Michigan Institute for Data Science (MIDAS). He obtained his PhD under the supervision of Peter Bartlett at the University of California at Berkeley. His research interests lie in machine learning including statistical learning theory, online learning, reinforcement learning and control theory, network analysis, and optimization for machine learning. He collaborates with scientists to seek novel applications of machine learning in mobile health, learning analytics, and computational chemistry. His research has been recognized with paper awards at COLT 2005, COLT 2011, and AISTATS 2015. He received an NSF CAREER award in 2015, a Sloan Research Fellowship in 2017, and an Adobe Data Science Research Award in 2020.

Join Zoom Meeting https://umich.zoom.us/j/97598571292

Meeting ID: 975 9857 1292

Passcode: XXXXXX (Will be sent via email to attendees)