Shaddin Dughmi

Shaddin Dughmi is an Associate Professor in the Department of Computer Science at the University of Southern California, affiliated with the Viterbi School of Engineering and the Theory Group. He holds a B.S. from Cornell University (2004) and a Ph.D. from Stanford University (2011). His research focuses on algorithmic techniques with applications to game theory, mechanism design, machine learning, and optimization. He has received prestigious awards including the NSF CAREER Award and the Arthur L. Samuel Thesis Award.

Education:

• B.S., Computer Science, Cornell University, 2004
• Ph.D., Computer Science, Stanford University, 2011

Research Interests: Algorithmic game theory, mechanism design, multi-agent systems, computational aspects of learning theory, submodular optimization, and approximation algorithms. His work bridges combinatorial and convex optimization techniques, with applications to information design and stochastic decision-making.

Recent Publications: Focus on foundational questions in learning theory (PAC learning, transductive learning), combinatorial optimization (matroid secretary, submodular functions), and algorithmic economics (delegation, stochastic selection). These works explore algorithmic boundaries and practical implementations across domains.

Awards:

• NSF CAREER Award
• Arthur L. Samuel Best Doctoral Thesis Award
• ACM EC Best Student Paper Award

Advising & Grants: Advised Ph.D. students including Tyler Labonte (now at Georgia Tech) and Ehsan Emamjomeh-Zadeh (Facebook Research), and co-advises current students like Yusuf Hakan Kalayci. Active in grant-funded research through NSF and other agencies, focusing on algorithmic foundations and interdisciplinary applications.

Labs/Teams: Core member of the USC Theory Group, collaborating on projects in algorithm design, computational economics, and machine learning. Teaches graduate courses in approximation algorithms, convex optimization, and discrete methods.