MIT-Harvard Communications Information Networks Circuits and Signals (CINCS) / Hamilton Institute Seminar

Wednesday, May 19, 2021 10:00am EDT

I'm Interested
Image of MIT-Harvard Communications Information Networks Circuits and Signals (CINCS) / Hamilton Institute Seminar
I'm Interested

About this Event

Virtual Event View stream information

Title: Learning Optimal Resource Allocations in Wireless Communication Systems

Abstract: Optimal resource allocation problems in wireless communication systems are mathematically equivalent to constrained statistical risk minimization problems. We show these nonconvex problems have O(e) duality gap if the learning parametrization is e-universal. This property motivates the development of primal-dual training methods that we show can outperform established heuristics. To enable scalability to networks with large numbers of transceivers we use graph neural networks (GNNs) as learning parametrizations. We explore stability and transferability properties of GNNs to explain their good empirical performance.

Wireless communication is a prototypical example of constrained learning and wireless networks are prototypical examples of distributed collaborative systems. We close the talk with a brief account of connections.

 

Zoom: https://mit.zoom.us/j/94416668228?pwd=LzJCV1VveUZ6WHdkQnlwektrVVkwQT09

Passcode: 825980

Event Details

Event Type

Colloquia / Seminar / Lecture

Academics

Computer Science Electrical & Computer Engineering

Website

https://www.rle.mit.edu/ncrc/mit-harv...

Speaker 1

Alejandro Ribeiro

Speaker Bio

Alejandro Ribeiro, received the B.Sc. degree in electrical engineering from the Universidad de la República Oriental del Uruguay in 1998 and the M.Sc. and Ph.D. degrees in electrical engineering from the Department of Electrical and Computer Engineering at the University of Minnesota in 2005 and 2007. He joined the University of Pennsylvania (Penn) in 2008 where he is currently Professor of Electrical and Systems Engineering. His research is in the applications of statistical signal processing to collaborative intelligent systems. His specific interests are in wireless autonomous networks, machine learning on network data and distributed collaborative learning. Papers coauthored by Dr. Ribeiro received the 2014 O. Hugo Schuck best paper award, and paper awards at ICASSP 2020, EUSIPCO 2019, CDC 2017, SSP Workshop 2016, SAM Workshop 2016, Asilomar SSC Conference 2015, ACC 2013, ICASSP 2006, and ICASSP 2005. His teaching has been recognized with the 2017 Lindback award for distinguished teaching and the 2012 S. Reid Warren, Jr. Award presented by Penn’s undergraduate student body for outstanding teaching. Dr. Ribeiro is a Fulbright scholar class of 2003 and Penn Fellow class of 2015.

Web: alelab.seas.upenn.edu.

Host

Prof. Flavio Clamon

Contact

Molly Kruko

Email

mkruko@mit.edu

See Who Is Interested

0 people are interested in this event

I'm Interested

Dial-In Information

Zoom: https://mit.zoom.us/j/94416668228?pwd=LzJCV1VveUZ6WHdkQnlwektrVVkwQT09

Passcode: 825980