Review for NeurIPS paper: Contextual Reserve Price Optimization in Auctions via Mixed Integer Programming

Summary and Contributions: In the second-price auction with reserve price, the publisher sets a reserve price before the auction, and the payment of the winner is defined as the maximum of the second highest bid and the reserve price. The paper deals with the problem of deciding the reserve price from the side information to maximize the payment from the winner. It considers using linear regression to infer the best reserve price. Then the problem is reduced to the optimization problem of optimizing the model parameters so as to maximize the payment. The main contribution of this paper is to discuss how to solve this optimization problem.