Group for Research in Decision Analysis

Exploiting degeneracy in minimax combinatorial optimization problems: applications to clustering, location analysis and graph partitioning

Claudio Contardo Associate Professor, Department of Management and Technology, Université du Québec à Montréal, Canada

Classical optimization tools as simplex-based enumeration tend to behave poorly in minimax combinatorial optimization problems due to degeneracy. I will discuss the use of two tools to handle these classes of problems: binary search and sampling. I will exemplify their use in two problems: the minimax diameter clustering problem and the p-center problem. Applications to graph partitioning will be briefly discussed.

Coffee and biscuits will be offered at the beginning of the seminar.
Welcome to everyone!