Stats staff seminar
Marian Scott and Xiaofei Zhang (University of Glasgow)
Wednesday 17th January 13:00-14:00 Maths 311B
Abstract
Science advice-helping inform policy. What does a statistician know? (Marian)
I will briefly talk about my involvement in a number of science committees and what my role is/has been.
Personalized Federated Learning with Fused and Sparse Penalization. (Xiaofei)
Federated learning (FL) is an emerging topic due to its advantage in collaborative learning with distributed data. Due to the heterogeneity in the local data-generating mechanism, it is important to consider personalization when developing federated learning methods. In this work, we propose a personalized federated learning (PFL) method to address the robust regression problem. First, to increase the estimation robustness, we aim to learn the regression weight by solving a Huber loss with the sparse fused penalty. We designed our personalized federated learning for robust and sparse regression algorithm to solve the estimation problem in the federated system with a central server efficiently. Furthermore, we extend the regression problem to empirical risk minimization. We also designed an algorithm for learning the empirical risk with fused penalization in a federated system over a decentralized network.
Add to your calendar
Download event information as iCalendar file (only this event)