Columbia Researchers Receive National Science Foundation + Amazon Grant

March 30, 2022

Five researchers from Columbia Engineering’s industrial engineering and operations research department have won a grant from the National Science Foundation Program on Fairness in Artificial Intelligence in Collaboration with Amazon to design algorithms for making fair decisions in AI-mediated auctions, pricing, and marketing.

The team includes Adam Elmachtoub, an associate professor of industrial engineering and operations research and an affiliated member of the Data Science Institute’s (DSI) Financial and Business Analytics, Foundations of Data Science, and Smart Cities centers; Shipra Agrawal, the Cyrus Derman Associate Professor of Industrial Engineering and Operations Research and an affiliate member of DSI’s Foundations of Data Science center; Eric Balkanski, an assistant professor of industrial engineering and operations research and an affiliated member of DSI’s Foundations of Data Science center; Rachel Cummings, an assistant professor of industrial engineering and operations research and an affiliated member of DSI’s Cybersecurity and Foundations of Data Science centers; and Christian Kroer, an assistant professor of industrial engineering and operations research and an affiliated member of DSI’s Cybersecurity, Foundations of Data Science, and Smart Cities centers.

FROM THE AWARD ABSTRACT

The deployment of AI systems in business settings has thrived due to direct access to consumer data, the capability to implement personalization, and the ability to run algorithms in real-time. For example, advertisements users see are personalized since advertisers are willing to bid more in ad display auctions to reach users with particular demographic features. Pricing decisions on ride-sharing platforms or interest rates on loans are customized to the consumer’s characteristics in order to maximize profit. Marketing campaigns on social media platforms target users based on the ability to predict who they will be able to influence in their social network. Unfortunately, these applications exhibit discrimination. Discriminatory targeting in housing and job ad auctions, discriminatory pricing for loans and ride-hailing services, and disparate treatment of social network users by marketing campaigns to exclude certain protected groups have been exposed. This project will develop theoretical frameworks and AI algorithms that ensure consumers from protected groups are not harmfully discriminated against in these settings. The new algorithms will facilitate fair conduct of business in these applications. The project also supports conferences that bring together practitioners, policymakers, and academics to discuss the integration of fair AI algorithms into law and practice.

The project develops novel theoretical frameworks to analyze algorithms according to both fairness and business objectives for three canonical business domains: auctions, pricing, and marketing. The approach considers three aspects of the decision-making pipeline. First, the project aims to understand the new types of criteria required to ensure fair auctions, pricing, and marketing, and designs novel algorithms that can incorporate these fairness criteria in real-world large-scale systems. Second, for each of these business contexts, the project considers how data can and should be collected in order to induce fair outcomes in the downstream decision-making task. Thirdly, the project considers how incorporating fairness measures, or failing to do so, can positively or negatively affect firms and consumers in the long-term, particularly in the presence of competition.