Inequality in Congestion Games with Learning Agents

2026-01-28

AI Summary

This study addresses how infrastructure expansion in transportation networks, while potentially enhancing overall efficiency, may exacerbate inequality due to heterogeneous learning capabilities among commuters. Modeling travelers as reinforcement learning agents with varying learning rates, the authors employ multi-agent simulations and congestion game analysis on both a Braess paradox network and an abstracted Amsterdam metro model to uncover the trade-off between efficiency and fairness during dynamic adaptation. The work introduces a novel metric—the “Price of Learning”—to quantify efficiency losses incurred during the learning phase, moving beyond traditional equilibrium-focused analyses. By highlighting how disparities in learning dynamics affect equitable outcomes, the study underscores the necessity for transport policies to account for heterogeneity in users’ adaptive behaviors.

Technology Category

• Search and Optimization: Learning to Search
• Planning, Routing, and Scheduling: Planning/Scheduling and Learning
• Game Theory and Economic Paradigms: Adversarial Learning

Application Category

• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Who benefits from expanding transport networks? While designed to improve mobility, such interventions can also create inequality. In this paper, we show that disparities arise not only from the structure of the network itself but also from differences in how commuters adapt to it. We model commuters as reinforcement learning agents who adapt their travel choices at different learning rates, reflecting unequal access to resources and information. To capture potential efficiency-fairness tradeoffs, we introduce the Price of Learning (PoL), a measure of inefficiency during learning. We analyze both a stylized network -- inspired in the well-known Braess's paradox, yet with two-source nodes -- and an abstraction of a real-world metro system (Amsterdam). Our simulations show that network expansions can simultaneously increase efficiency and amplify inequality, especially when faster learners disproportionately benefit from new routes before others adapt. These results highlight that transport policies must account not only for equilibrium outcomes but also for the heterogeneous ways commuters adapt, since both shape the balance between efficiency and fairness.

Problem

• congestion games
• inequality
• transport networks
• reinforcement learning
• efficiency-fairness tradeoff

Innovation

• Price of Learning
• reinforcement learning agents
• congestion games
• transport inequality
• heterogeneous adaptation