Topic: Privacy-Preserving Statistical Inference and Machine Learning with Differential Privacy

Abstract:

In today's world, our diverse information is collected through various channels and utilized for a range of purposes, including statistical inference and the development of machine learning models. However, privacy threats continue to emerge, revealing that sensitive personal information can be inferred from statistics or machine learning models. To address this challenge, appropriate privacy-preserving mechanisms must be integrated into data collection and utilization. These lectures focus on differential privacy, a fundamental privacy notion, and its applications in privacy-preserving statistical inference and machine learning. We will explore its theoretical underpinnings, practical implementations, and cutting-edge research directions. These lectures are structured as follows.

1) The first lecture will introduce privacy risks in statistical inference and machine learning, followed by a rigorous definition of differential privacy and an examination of its key properties.

2) The second lecture will focus on the realization of differential privacy in machine learning, with a focus on differentially private training.

3) The third lecture will investigate the optimal privacy-utility trade-off in statistical inference, particularly in the context of distribution estimation. Specifically, we will examine how classical results in combinatorial design can be effectively leveraged to achieve this trade-off in a communication-efficient manner.

4) As quantum technology continues to advance and is expected to become more widespread, a key question that must first be addressed when applying quantum technology to specific systems is whether it would bring a fundamental quantum advantage. In the last lecture, we will explore how quantum technology can improve the fundamental privacy-utility trade-off in statistical inference.

Bio:

Prof. Si-Hyeon Lee received the B.S. (summa cum laude) and Ph.D. degrees in electrical engineering from the Korea Advanced Institute of Science and Technology (KAIST), Daejeon, South Korea, in 2007 and 2013, respectively. She is currently an Associate Professor with the School of Electrical Engineering, KAIST. She was a Postdoctoral Fellow with the Department of Electrical and Computer Engineering, University of Toronto, Toronto, Canada, from 2014 to 2016, and an Assistant Professor with the Department of Electrical Engineering, Pohang University of Science and Technology (POSTECH), Pohang, South Korea, from 2017 to 2020. Her research interests include information theory, wireless communications, statistical inference, and machine learning. She is currently an IEEE Information Theory Society Distinguished Lecturer and serving as an Associate Editor for the IEEE Transactions on Information Theory.