Robust Dynamic Pricing with Demand Learning in the Presence of Outlier Customers
研究了存在异常顾客时动态定价问题，提出能应对任意异常到达和需求模式的稳健定价策略，无需事先知道异常比例。
Dynamic pricing is a core problem in revenue management. Most existing literature assumes that the demand follows a probabilistic model, with an unknown demand curve as the mean. However, in practice, customers may not always behave according to such a model. In “Robust Dynamic Pricing with Demand Learning in the Presence of Outlier Customers,” Chen and Wang study the dynamic pricing problem under model misspecification. To characterize the behavior of outlier customers, an ε-contamination model—the most fundamental model in robust statistics and machine learning, is adopted. The challenges brought by the presence of outlier customers are mainly due to the fact that arrivals of outliers and their exhibited demand behaviors are completely arbitrary. To address these challenges, the authors propose robust dynamic pricing policies that can handle any outlier arrival and demand patterns. The proposed policies are fully adaptive without requiring prior knowledge of the outlier proportion parameter.