Optimizing Platforms for Multiple Objectives with ML

Talk:

• Multi-sided platforms have witnessed an explosive growth by facilitating efficient interactions between multiple stakeholders, including e.g. buyers and retailers (Amazon), guests and hosts (AirBnb), riders and drivers (Uber), and listeners and artists (Spotify). A large number of such platforms rely on machine learning powered matching engines connecting consumers with suppliers by acting as a central platform, thereby finding the right fit and efficiently mediating interactions between the two sides.
• In this talk we discuss a number of problems which need to be addressed when developing a search & recommendation framework powering multi-stakeholder platforms. We begin by describing a contextual bandit model developed for serving explainable music recommendations to users and showcase the need for explicitly considering supplier-centric objectives during optimization.
• We highlight the importance of a multi-objective ranking/recommendation and discuss different ways in which stakeholders specify their objectives. Finally, we demonstrate how enhanced user and content understanding helps us in developing better models to power multi-stakeholder platforms.