Three Snippets of Taming the Nonconvex: QCQP, MIP, and MSIP
Add to Google Calendar
In this talk, I will report some recent advances in handling nonconvexity in three diverse classes of problems: nonconvex QCQP, mixed integer linear programming (MILP), and multistage stochastic integer programming (MSIP). The methodological innovation includes strong SOCP relaxations of quadratic constraints, strong augmented Lagrangian duality for MILP, and strong Lagrangian duality for binary linear programs. These techniques will be illustrated by fundamental optimization problems in electric energy systems.
Andy Sun is an assistant professor in the H. Milton Stewart School of Industrial and Systems Engineering at Georgia Institute of Technology. He obtained the doctoral degree from the Operations Research Center at MIT. He has broad research interest in optimization under uncertainty, and nonconvex and convex optimization, with applications in electric energy systems. His research won several awards including the second prize in George Dantzig Dissertation Award in 2011, INFORMS ENRE Best Paper in Energy in 2017, and the NSF CAREER award.