The popular assignment problem: when cardinality is more important than popularity
Nitya Raju - University of Maryland, College Park
IRB 3137 or Zoom: https://umd.zoom.us/j/6778156199?pwd=NkJKZG1Ib2Jxbmd5ZzNrVVlNMm91QT09
Abstract
In this talk I will present a paper by Telikepalli Kavitha, Tamás Király, Jannik Matuschke, Ildikó Schlotter, Ulrike Schmidt-Kraepelin on the popular assignment problem. The popular matching problem is a variant of the classical stable matching problem, and we consider the house allocation setting where we would like to match houses to agents who have preferences over the houses. We will see a nice characterization of popular matchings in the case of strict preferences, and then the primal-dual algorithm given by the authors that resolves the problem for more general preferences.
Bio
Nitya is a second year PhD student at the University of Maryland advised by Aravind Srinivasan, working on the design and analysis of algorithms. She is interested in popular and stable matchings in addition to working on randomized algorithms.
This talk is organized by Kishen N Gowda