Presentation details:

Title: Learning to Price: From Samples to Queries

Date: August 27th, 2025

Time: 11:00 AM – 12:00 PM EDT

Location: KACB 3100 

Committee:

Prof. Sahil Singla (Advisor), School of Computer Science, Georgia Institute of Technology

Prof. Jan van den Brand, School of Computer Science, Georgia Institute of Technology

Prof. Santosh Vempala, School of Computer Science, Georgia Institute of Technology

Prof. Will Ma, Graduate School of Business and Data Science Institute, Columbia University

Prof. Rad Niazadeh, Booth School of Business, University of Chicago

Abstract:

In many everyday transactions, sellers set fixed prices for their goods, and buyers decide whether to purchase based on those prices. This simple process, known as posted pricing, captures the essence of many real-world markets: a seller tags an item with a price, and buyers choose their preferred bundles that maximize personal utility. A standard assumption in models of this process is that buyers’ valuations for items are drawn from underlying distributions. The seller’s challenge is to learn enough about these distributions—through data or interactions with buyers—in order to set prices that achieve desirable outcomes such as maximizing social welfare or revenue.

This proposal investigates the fundamental question of how much information is needed to design effective posted pricing mechanisms when buyers’ valuation distributions are unknown. I study this question under two complementary models of information: the sample model, where the seller can observe data points drawn from buyers’ distributions, and the query model, where the seller only observes which bundle a buyer selects given the current posted prices. My results provide characterizations of the sample and query complexity of posted pricing in several key settings. These findings contribute to a broader understanding of the learnability of market mechanisms under realistic informational constraints, and they provide insights that can guide the design of practical mechanisms in real-world markets.