Federated Learning for Nonparametric Function Estimation: Framework and Optimality
Abstract
Federated learning is a machine learning paradigm designed to tackle the challenges of data governance and privacy. It enables organizations (e.g., hospitals) to collaboratively train and enhance a shared global statistical model without sharing raw data externally. Instead, the learning process occurs locally at each participating entity, and only model characteristics, such as parameters and gradients, are exchanged, while preserving privacy.
In this talk, the speaker and his research group consider statistical optimality for federated learning in the context of nonparametric regression. The setting they study is heterogeneous, encompassing varying sample sizes and differential privacy constraints across different servers. Within this framework, both global and pointwise estimation are considered, and optimal rates of convergence over the Besov spaces are established.
The speaker and his research group propose distributed privacy-preserving estimation procedures and analyze their theoretical properties. The findings shed light on the delicate balance between accuracy and privacy preservation. In particular, they characterize the compromise not only in terms of the privacy budget but also concerning the loss incurred by distributing data within the privacy framework as a whole. This insight captures the folklore wisdom that it is easier to retain privacy in larger samples, and explores the differences between pointwise and global estimation under distributed privacy constraints.
About the Speaker
Prof. Tony Cai received his PhD from Cornell University in 1996. He joined the University of Pennsylvania in 2006 and is currently the Daniel H. Silberberg Professor and Professor of Statistics and Data Science. He was also named the Medallion Lecturer at the Institute of Mathematical Statistics in 2009.
Prof. Cai’s research interests include high-dimensional inference, large-scale multiple testing, nonparametric function estimation, functional data analysis, inference for discrete distributions, and statistical decision theory, with applications to compressed sensing, chemical identification, medical imaging and microarray data analysis. He currently serves on the editorial board of the Frontiers of Statistics (book series). He was also the editor of The Annals of Statistics in 2010-2012.
Prof. Cai received the Committee of Presidents of Statistical Societies (COPSS) Presidents' Award in 2008, the Noether Distinguished Scholar Award from the American Statistical Association in 2023 and the Frontiers of Science Award at the 2023 International Congress of Basic Science. He was elected as a Fellow of the Institute of Mathematical Statistics in 2006, the President of International Chinese Statistical Association in 2016, and a Fellow of American Association for the Advancement of Science in 2024.
For Attendees' Attention
Seating is on a first come, first served basis.
