Harvard Electrical Engineering Seminar

Friday, April 24 | 11am - 12pm

SEC LL2.224

"Wireless Federated Learning"

Speaker: Vincent Poor
Michael Henry Strater University Professor, Princeton University

Abstract: Federated learning is a distributed machine learning technique in which multiple learning agents collaborate to learn a common model using individual datasets, often with the help of an aggregator.  Federated learning is a natural approach for learning in mobile and other wireless networks, such as those supporting autonomous vehicles or factory automation, in which data is typically collected at the network edge, and latency and bandwidth constraints limit the degree to which such data can be centralized for learning and inferential purposes.   Such application introduces the issue of how the wireless medium interacts with the machine learning algorithms, including how the vagaries of wireless transmission affect learning performance, how agents’ participation in model aggregation should be scheduled to make optimal use of wireless resources, and how to mitigate privacy and security issues, which are exacerbated by the open nature of the wireless medium.  This talk will provide a brief introduction to this field and discuss progress in addressing the above and other issues arising therein.

Speaker Bio: H. Vincent Poor is the Michael Henry Strater University Professor at Princeton University, where his interests include information theory, stochastic analysis and machine learning, and their applications in wireless networks, energy systems and related fields. Among his publications in these areas is the book Machine Learning and Wireless Communications, published by Cambridge University Press. Dr. Poor is a member of the National Academy of Engineering and the National Academy of Sciences, and he is a foreign member of the Royal Society and other national and international academies.  Other recognition of his work includes the IEEE Alexander Graham Bell Medal and honorary doctorates from a number of universities in Asia, Europe and North America.