Fair-GNE : Generalized Nash Equilibrium-Seeking Fairness in Multiagent Healthcare Automation

2025-11-17

AI Summary

Existing multi-agent reinforcement learning (MARL) approaches for demand-side automation in healthcare suffer from unfair workload allocation, relying on post-hoc reward shaping without runtime-enforceable, self-verifiable fairness guarantees. Method: We propose Fair-GNE, the first framework to explicitly encode fairness as a hard constraint within a generalized Nash equilibrium (GNE), ensuring runtime self-enforcement and immutability. Integrating constrained optimization with game-theoretic modeling, we design an adaptive constraint enforcement algorithm that computes locally efficient and safety-aware equilibrium policies in a high-fidelity resuscitation simulator. Results: Experiments demonstrate a Jain fairness index of 0.89—significantly surpassing baselines (0.33, *p* < 0.01)—while maintaining an 86% task success rate, validating Fair-GNE’s effectiveness and practicality in large-scale clinical systems.

Technology Category

• Game Theory and Economic Paradigms: Fair Division
• Multiagent Systems: Mechanism Design
• Philosophy and Ethics of AI: Bias, Fairness & Equity

Application Category

• Economics, Online Markets and Human Computation: Fairness and ethical considerations in crowd work and in human-in-the-loop AI systems
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Enforcing a fair workload allocation among multiple agents tasked to achieve an objective in learning enabled demand side healthcare worker settings is crucial for consistent and reliable performance at runtime. Existing multi-agent reinforcement learning (MARL) approaches steer fairness by shaping reward through post hoc orchestrations, leaving no certifiable self-enforceable fairness that is immutable by individual agents at runtime. Contextualized within a setting where each agent shares resources with others, we address this shortcoming with a learning enabled optimization scheme among self-interested decision makers whose individual actions affect those of other agents. This extends the problem to a generalized Nash equilibrium (GNE) game-theoretic framework where we steer group policy to a safe and locally efficient equilibrium, so that no agent can improve its utility function by unilaterally changing its decisions. Fair-GNE models MARL as a constrained generalized Nash equilibrium-seeking (GNE) game, prescribing an ideal equitable collective equilibrium within the problem's natural fabric. Our hypothesis is rigorously evaluated in our custom-designed high-fidelity resuscitation simulator. Across all our numerical experiments, Fair-GNE achieves significant improvement in workload balance over fixed-penalty baselines (0.89 vs. 0.33 JFI, $p < 0.01$) while maintaining 86% task success, demonstrating statistically significant fairness gains through adaptive constraint enforcement. Our results communicate our formulations, evaluation metrics, and equilibrium-seeking innovations in large multi-agent learning-based healthcare systems with clarity and principled fairness enforcement.

Problem

• Achieving fair workload allocation in multiagent healthcare systems
• Addressing runtime fairness through constrained generalized Nash equilibrium
• Ensuring equitable resource sharing among self-interested healthcare agents

Innovation

• Models MARL as constrained generalized Nash equilibrium game
• Uses adaptive constraint enforcement for workload balance
• Achieves equitable collective equilibrium in multiagent systems