Discipline: Computer/Information Sciences
My research aims to settle the complexity status of central problems in combinatorial optimization, such as the Unique Games problem and Sparsest Cut. Candidate methods are poorly understood semidefinite programming hierarchies. For any significant progress a deeper understanding of the underlying geometry has to be developed, bridging several areas of mathematics and theoretical computer science.