IDEaS Short Talks Presents Justin Romberg

Event Details
  • Date/Time:
    • Monday November 19, 2018
      2:00 pm - 2:30 pm
  • Location: TSRB Auditorium
  • Phone:
  • URL:
  • Email:
  • Fee(s):
  • Extras:
    Free food


Summary Sentence: Justin Romberg talks about “Solving (linear and nonlinear) Systems of Equations using Convex Programming"

Full Summary: No summary paragraph submitted.

  • ECE Assistant Professor Justin Romberg ECE Assistant Professor Justin Romberg
IDEaS Short Talks and Networking Social
“Solving (linear and nonlinear) Systems of Equations using Convex Programming" by Justin Romberg
Monday, November 19, 2018
Part of One Short 30-MinuteTalks from 2-3 pm
Networking Social from 3-3:30 pm
Technology Square Research Building (TSRB) Auditorium
(You are welcome to attend any part of these events as your schedule permits.)
IDEaS is running a series of short talks to learn about research across the Georgia Tech campus. The presentations are from broadly different topics and accessible to those in other research areas. Come chat with colleagues and learn more about what’s new in Data Science! 
Justin Romberg
Professor in the School of Electrical and Computer Engineer
Georgia Institute of Technology
2:00 - 2:30 pm

“Solving (linear and nonlinear) Systems of Equations using Convex Programming" 

Abstract: Solving linear systems of equations, Ax=b, is the most fundamental problem in applied mathematics. Over the past 15 years, our understanding of when and how this linear inverse problem can be solved has evolved dramatically.  In particular, we know that under determined systems of equations can be solved robustly if their solution is structured.  How certain kinds of structure helps in solving linear systems of equations is completely understood through convex programming: analysis of when systems have unique solutions and robustness can be translated to simple geometrical concepts.

The geometrical concepts and analytical techniques used for finding structured solutions to linear systems also give us insight into how and when nonlinear systems of equations can be solved.  Solving these systems by minimizing a loss function (what is called "empirical risk minimization" in machine learning), typically requires solving a nonconvex optimization program.  We will discuss a very general (and very natural) convex relaxation for solving nonlinear systems both with and without structure.
Bio: Dr. Justin Romberg is the Schlumberger Professor in the School of Electrical and Computer Engineering at the Georgia Institute of Technology, where he has been on the faculty since 2006.  Dr. Romberg received the B.S.E.E. (1997), M.S. (1999) and Ph.D. (2004) degrees from Rice University in Houston, Texas; in 2010, he was named a Rice University Outstanding Young Engineering Alumnus.  From Fall 2003 until Fall 2006, he was a Postdoctoral Scholar in Applied and Computational Mathematics at the California Institute of Technology. He is currently on the editorial board for the SIAM Journal on the Mathematics of Data Science, and is a Fellow of the IEEE.  His current research interests lie at the intersection of statistical signal processing, machine learning, optimization, and applied probability.


Additional Information

In Campus Calendar

Institute for Data Engineering and Science

Invited Audience
Faculty/Staff, Postdoc, Public, Graduate students, Undergraduate students
No categories were selected.
machine learning, data science, Electrical and Computer Engineering
  • Created By: shalbert6
  • Workflow Status: Published
  • Created On: Nov 9, 2018 - 3:20pm
  • Last Updated: Nov 9, 2018 - 3:48pm