Description:

Monday / CS Seminar
November 17th/ 12:00 PM
Hybrid / Mudd 3514

Speaker
Shaddin Dughmi, USC

Talk Title
A Combinatorial Lens on Supervised Learning

Abstract
I will describe some recent joint results on the theory of supervised learning which follow from surprising connections to combinatorial optimization. Most notably, by connecting learning to bipartite matching on infinite graphs we derive several algorithmic and structural characterizations of learning which hold somewhat broadly.

Biography
"Shaddin Dughmi is an Associate Professor in the Department of Computer Science at USC and the Neild visiting professor at Northwestern University. He received a B.S. in computer science, summa cum laude, from Cornell University in 2004, and a Ph.D. in computer science from Stanford University in 2011. He is a recipient of the NSF CAREER award, the Arthur L. Samuel best doctoral thesis award, and the ACM EC best student paper award. 
Shaddin is broadly interested in questions that stimulate the development of new algorithmic techniques, and shed insight on the power and limitations of algorithms. Work in his group has investigated such questions in a variety of domains: game theory, mechanism design, multi-agent systems, persuasion and information design, delegation and contract theory, decision making subject to online or stochastic uncertainty, and the theory of machine learning."

---

CSPAC Community Connections:
Zoom Link
Panopto Link