Etan Green

My current research applies deep reinforcement learning to real-world economic problems, such as bargaining and pricing. I am broadly interested in optimal behavior in complex decision-making environments, and how human behavior compares.

» etangr [at] upenn.edu

Bio

2021-present: Visiting Researcher, Arena-AI
2016-present: Assistant Professor of Operations, Information and Decisions, The Wharton School, University of Pennsylvania
2015-16: Postdoc Researcher, Microsoft Research, New York City
2015: Ph.D., Stanford University Graduate School of Business
2008-10: Associate Consultant, Bain & Company, San Francisco
2008: A.B., Brown University

Papers

Optimal Bargaining on eBay Using Reinforcement Learning, Etan Green & Barry Plunkett. Under review.
» An artificial intelligence that provides simple and effective strategies for bargaining in one of the world’s largest markets.
» A method for applying deep reinforcement learning to real-world economic problems.
paper | appendix | talk | github: rl | github: data processing


The Favorite-Longshot Midas, Etan Green, Haksoo Lee & David Rothschild. Invited revision at Review of Economic Studies.
» Racetracks dupe bettors into overbetting longshots—and make more money as a result.
paper


Bayesian Instinct, Etan Green & David Daniels. Under review.
» Umpires are not biased—they're Bayesian.
paper | talk | github


A Sharp Test of the Portability of Expertise, Etan Green, Justin Rao & David Rothschild. Management Science, 2019.
» Expertise is not very portable.
paper | supplement + data & code


Personal Bests as Reference Points, Ashton Anderson & Etan Green. Proceedings of the National Academy of Sciences, 2018.
» People care a lot about achieving their best-ever performance.
paper | appendix | data & code


What Does it Take to Call a Strike? Three Biases in Umpire Decision Making, Etan Green & David Daniels. MIT Sloan Sports Analytics Conference, 2014.
» Umpires are biased.
paper | talk


Based on a design by Ashton Anderson.