Itai Gurvich

Professor
Cornell

Abstract: In this talk, I will discuss a family of dynamic resource allocation problems that includes, as specific instances, network revenue management and dynamic pricing. I will show that the value of information (or the regret) -- the expected gap in performance of the best online (sequential) algorithm and its offline (full information) counterpart---is bounded irrespective of the horizon length or the initial inventory. This bounded regret is achieved by tractable algorithms that resolve a linear program at every period but very carefully translate the LP solution into dynamic actions. These tractable policies, have, in turn, a bounded optimality gap. Our approach to the proof is one that relates the geometry of a suitable packing LP to the dynamics of the residual-inventory process. 

Bio: Itai Gurvich is a Professor at Cornell Tech and in the Operations Research and Information Engineering Department at Cornell University. He earned a Ph.D. from the Decision, Risk and Operations department at Columbia University’s Graduate School of Business. He spent 8 years teaching at the Kellogg School of Management at Northwestern University. His research interests include performance analysis and optimization of processing networks, the theory of stochastic-process approximation and the application of operations research and statistical tools to healthcare processes.