​Most interesting optimization problems are NP-hard, which means that polynomial time algorithms that find the exact optimal solution are unlikely to exist. "Approximation algorithms" provide the most natural ways to deal with this issue by finding an approximate optimal solution in polynomial time.

Continuous and discrete optimization are the two main branches of mathematical optimization that have been primarily studied in different contexts. While the theory of approximate algorithms for discrete optimization has been widely successful, its counterpart for continuous optimization has many basic questions answered. In this talk, we will discuss some connections between discrete and continuous optimizations that result in an exciting but not complete theory of approximate continuous optimization. No background in approximation algorithms will be assumed.

Euiwoong Lee is an assistant professor at the University of Michigan. He got his Ph.D. from Carnegie Mellon University in 2017 and worked as a research fellow at the Simons Institute for the Theory of Computing at UC Berkeley and a postdoc at New York University. His research interests lie in several topics of theoretical computer science including approximation algorithms and hardness of approximation. His work has been recognized by several awards, including NSF CAREER Award, Edmund M. Clarke Dissertation Award, and ICALP Best Student Paper Award.