A Market for Accuracy: Classification Under Competition

Ohad Einav, Nir Rosenfeld

TL;DR: We study learning in a setting where multiple classifiers compete over users in a `market for accuracy'; we propose an effective approach for learning under competition and analyze market outcomes

Abstract: Machine learning models play a key role for service providers looking to gain market share in consumer markets. However, traditional learning approaches do not take into account the existence of additional providers, who compete with each other for consumers. Our work aims to study learning in this market setting, as it affects providers, consumers, and the market itself. We begin by analyzing such markets through the lens of the learning objective, and show that accuracy cannot be the only consideration. We then propose a method for classification under competition, so that a learner can maximize market share in the presence of competitors. We show that our approach benefits the providers as well as the consumers, and find that the timing of market entry and model updates can be crucial. We display the effectiveness of our approach across a range of domains, from simple distributions to noisy datasets, and show that the market as a whole remains stable by converging quickly to an equilibrium.

Lay Summary: Machine learning models are widely used in consumer markets, but most ignore the fact that companies are in competition for the same users. We developed a new learning approach that helps models (i.e. service providers) compete more effectively by considering market dynamics. This can improve outcomes for both providers and consumers.

Primary Area: Theory->Game Theory

Keywords: accuracy markets, best-response classifiers, game theory, competition

Submission Number: 2589