Learning Classifiers That Induce Markets

2025-02-27

AI Summary

In strategic classification settings, where agents deliberately manipulate features to obtain favorable predictions, the standard exogenous cost assumption fails because classifier deployment endogenously induces a feature market—agents’ competition for critical features spontaneously generates market prices. Method: The paper introduces the “classifier-induced market” paradigm, the first framework to endogenize market mechanisms within strategic classification. It extends the strategic classification model to incorporate supply-demand dynamics, designs an algorithm to compute market equilibria, and develops an end-to-end differentiable learning framework. Contribution/Results: Theoretically, the paper proves existence of equilibrium and convergence of the proposed algorithm. Empirically, the method robustly induces economically plausible price structures and significantly improves social welfare compared to baseline approaches—demonstrating that explicitly modeling strategic feature markets enhances both fairness and efficiency in deployed classifiers.

Technology Category

• Game Theory and Economic Paradigms: Mechanism Design
• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Multiagent Systems: Mechanism Design

Application Category

• Economics, Online Markets and Human Computation: Uses of LLMs and GenAI for marketplace design, bidding, and strategic interactions
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

When learning is used to inform decisions about humans, such as for loans, hiring, or admissions, this can incentivize users to strategically modify their features to obtain positive predictions. A key assumption is that modifications are costly, and are governed by a cost function that is exogenous and predetermined. We challenge this assumption, and assert that the deployment of a classifier is what creates costs. Our idea is simple: when users seek positive predictions, this creates demand for important features; and if features are available for purchase, then a market will form, and competition will give rise to prices. We extend the strategic classification framework to support this notion, and study learning in a setting where a classifier can induce a market for features. We present an analysis of the learning task, devise an algorithm for computing market prices, propose a differentiable learning framework, and conduct experiments to explore our novel setting and approach.

Problem

• Classifiers influence market formation
• Strategic feature modification by users
• Algorithm computes induced market prices

Innovation

• Inducing feature markets
• Differentiable learning framework
• Algorithm for market pricing