Machine Learning for Strategic Inference

TL;DR

This paper explores how algorithms guiding strategic players can induce near-rational behavior in markets, using adaptive boosting techniques under certain learnability conditions, offering a statistical view on endogenous model misspecification.

Contribution

It introduces a framework where adaptive boosting algorithms can promote rational-like behavior in strategic interactions under limited information and decision rules.

Findings

Adaptive boosting can induce near-rational behavior.

Weak learnability is key for algorithm effectiveness.

Provides a statistical perspective on model misspecification.

Abstract

We study interactions between strategic players and markets whose behavior is guided by an algorithm. Algorithms use data from prior interactions and a limited set of decision rules to prescribe actions. While as-if rational play need not emerge if the algorithm is constrained, it is possible to guide behavior across a rich set of possible environments using limited details. Provided a condition known as weak learnability holds, Adaptive Boosting algorithms can be specified to induce behavior that is (approximately) as-if rational. Our analysis provides a statistical perspective on the study of endogenous model misspecification.