Inverse Optimization of Stabilizing Grand Coalitions via Cost Vector Adjustment

2018-12-05

 TopicInverse Optimization of Stabilizing Grand Coalitions via Cost Vector Adjustment

Speaker: LIU Lindong

Time:  3:00pm, Monday,  December 10,2018

Site: EMS A208

HostXU Minghui

AbstractIn this paper, we propose an instrument for stabilizing the grand coalitions in unbalanced cooperative games via cost vector adjustment (CVA). To study the CVA instrument, we define its associating problem as a constrained inverse optimization problem (CIOP), and then formulate the optimization problem as a linear programming. We investigate the sufficient, the necessary and the sufficient and necessary conditions under which the CIOP is feasible: in the infeasible case, suggest to unify the CVA instrument with some existing instruments for stabilizing the grand coalitions; while in the feasible case, we study the computational complexity of the CIOP and propose two types of solution methods, namely, the cutting plane and the cone optimization methods, to solve it. In the end, we demonstrate the applicability of the CVA instrument to the weighted matching and uncapacitated facility location games.

 

Introduction to the Speaker

Since 2016, Dr. Liu Lindong has been an associate professor in School of Management, University of Science and Technology of China. In 2015, he received his doctorate in industrial engineering and decision analysis from the Hong Kong University of Science and Technology. In 2011, he received his bachelor's degree in automation from School of Engineering Management in Nanjing University. His primary research field is  cooperative gaming, and his findings have been published in Operations Research, INFORMS Journal on Computing and other UTD 24 top business journals.