Linde Institute/SISL Seminar

ABSTRACT: In many online marketplaces and service systems, there is an informational asymmetry between the platform operators and the users of the system: users often have much less information about system state than is available to the operators. Since state information directly affects user behavior, a natural question is how should a platform share information with users in order to achieve system-wide goals. In this talk, I study this question using the methodology of Bayesian persuasion in two operational contexts. First, I consider the problem of how a service provider should share queue-length information with customers in order to maximize throughput and revenue. Here, we establish structural results on the optimal signaling mechanism, and evaluate its revenue performance against state-dependent pricing mechanisms. Second, I consider the problem of demand and inventory signaling in a model of online retail, and study the revenue-performance of public signaling mechanisms. In contrast to standard models of Bayesian persuasion, we observe that both settings require careful assignment of agents' *prior* beliefs. I conclude with a brief discussion of information design in settings where agents exhibit risk-aversion.
(Joint work with David Lingenbrink, Cornell University.)

BIO: Krishnamurthy Iyer is an Assistant Professor in the School of Operations Research and Information Engineering at Cornell University. He received his PhD from the Department of Management Science and Engineering at Stanford University in 2012, and his B.Tech and M.Tech (dual degree) in Mechanical Engineering at the Indian Institute of Technology, Bombay in 2006. Before coming to Cornell, he was a Postdoctoral Researcher at the Computer and Information Science Department in the University of Pennsylvania. His research interests include game theory and mechanism design as applied to studying dynamic markets and service systems. His research is supported by grants from the National Science Foundation, and has received first place at the 2017 INFORMS Junior Faculty Paper Competition.