Two-Sided Pricing and Learning with Buyer Choice: A Generalized SGD Method

Learning to Price in Two-Sided Markets Many two-sided platforms set purchase prices for sellers and selling prices for buyers. In “Two-Sided Pricing and Learning with Buyer Choice: A Generalized SGD Method,” Lin and Huh study how platforms can make these decisions when responses on both sides are uncertain and must be learned from limited data. They first show that, when supply and demand models are known, simple fixed-price policies that match expected supply and demand can be asymptotically optimal. Building on this insight, the authors develop a novel generalization of classical stochastic gradient descent (SGD) for learning a fixed-price policy. Unlike standard SGD, their method does not require the construction of a strongly convex objective function. This flexibility allows it to accommodate demand interactions arising from buyer choice. The resulting learning algorithm is near-optimal when the planning horizon is large.