We study platforms in the sharing economy and discuss theneed for incentivizing users to explore options that otherwisewould not be chosen. For instance, rental platforms such asAirbnb typically rely on customer reviews to provide userswith relevant information about different options. Yet, often alarge fraction of options does not have any reviews available.Such options are frequently neglected as viable choices, andin turn are unlikely to be evaluated, creating a vicious cycle.Platforms can engage users to deviate from their preferredchoice by offering monetary incentives for choosing a differ-ent option instead. To efficiently learn the optimal incentivesto offer, we consider structural information in user prefer-ences and introduce a novel algorithm - Coordinated OnlineLearning (CoOL) - for learning with structural informationmodeled as convex constraints. We provide formal guaran-tees on the performance of our algorithm and test the viabil-ity of our approach in a user study with data of apartments onAirbnb. Our findings suggest that our approach is well-suitedto learn appropriate incentives and increase exploration on theinvestigated platform.

Top