Learning Optimal Posted Prices for a Unit-Demand Buyer

TL;DR

This paper investigates how to efficiently learn optimal item prices for a buyer with independent item values, using different query models, and provides nearly tight bounds on the required samples and queries.

Contribution

It introduces nearly tight bounds on sample and query complexity for learning optimal posted prices in a unit-demand setting under two common query models.

Findings

Nearly tight sample complexity bounds established.

Nearly tight pricing query complexity bounds established.

Applicable to practical pricing scenarios with limited buyer interaction.

Abstract

We study the problem of learning the optimal item pricing for a unit-demand buyer with independent item values, and the learner has query access to the buyer's value distributions. We consider two common query models in the literature: the sample access model where the learner can obtain a sample of each item value, and the pricing query model where the learner can set a price for an item and obtain a binary signal on whether the sampled value of the item is greater than our proposed price. In this work, we give nearly tight sample complexity and pricing query complexity of the unit-demand pricing problem.