Abstract
We consider a Markov persuasion process where a single long-lived sender persuades a stream of myopic receivers by sharing information about a payoff-relevant state. The state transitions are Markovian conditional on the receivers' actions, and the sender seeks to maximize the long-run average reward by committing to a (possibly history-dependent) signaling mechanism. Such problems are common in platform markets, where the platform seeks to influence users' actions to achieve desirable long-term revenue and welfare outcomes. To reflect the fact that in such settings the users may have partial information about the past, we analyze this problem under various information models that differ in the amount of information the receivers have about the history of the process. Specifically, in addition to the full-history information model (where the receivers observe the entire history) and the no-history information model (where they have no historical information), we analyze a partial-information model where each receiver observes the state-action pair $\ell$ periods ago (but not the subsequent transitions). We formulate the sender's problem under each information model as a linear program, and establish that the sender's payoff is the highest under the no-history information model. This implies that, in choosing the signaling mechanism, the sender faces a tradeoff between obtaining higher payoffs and being persuasive under a larger class of information models. Finally, we use a robustness framework to design a \emph{history-independent} signaling mechanism that achieves payoff close to that under the no-history information model while being persuasive even with some historical information.
Biography
Krishnamurthy Iyer is an Associate Professor in the Department of Industrial and Systems Engineering at the University of Minnesota. Previously, he was an Assistant Professor in the School of Operations Research and Information Engineering at Cornell University, and a Postdoctoral Researcher at the Computer and Information Science Department at the University of Pennsylvania. He received his PhD from the Department of Management Science and Engineering at Stanford University in 2012. His research interests include game theory, mechanism design and stochastic modeling, with applications in markets and service systems.
