MURIM: Multidimensional Reputation-based Incentive Mechanism for Federated Learning

2025-12-15

AI Summary

To address insufficient client incentives, privacy leakage, and malicious participation in federated learning, this paper proposes a multidimensional dynamic reputation-based incentive framework. Methodologically, it introduces the first integrated reputation verification mechanism and establishes a three-dimensional incentive model jointly considering contribution, latency, and reputation; designs a reliability verification module, a differential privacy enhancement mechanism, and a robust aggregation strategy; and proposes a multi-objective weighted incentive allocation algorithm. Key innovations include fine-grained identification of malicious clients and real-time interception of unfair rewards, ensuring stable model convergence under heterogeneous and dynamic environments. Experiments on MNIST, FMNIST, and ADULT demonstrate an 18% improvement in fairness, a 5–9% reduction in privacy attack success rate, and up to 85% enhanced robustness against poisoning and noise-gradient attacks.

Technology Category

• Machine Learning: Privacy
• Multiagent Systems: Mechanism Design
• Natural Language Processing: Safety and Robustness

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Security and Privacy: Large-scale security measurements
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

Federated Learning (FL) has emerged as a leading privacy-preserving machine learning paradigm, enabling participants to share model updates instead of raw data. However, FL continues to face key challenges, including weak client incentives, privacy risks, and resource constraints. Assessing client reliability is essential for fair incentive allocation and ensuring that each client's data contributes meaningfully to the global model. To this end, we propose MURIM, a MUlti-dimensional Reputation-based Incentive Mechanism that jointly considers client reliability, privacy, resource capacity, and fairness while preventing malicious or unreliable clients from earning undeserved rewards. MURIM allocates incentives based on client contribution, latency, and reputation, supported by a reliability verification module. Extensive experiments on MNIST, FMNIST, and ADULT Income datasets demonstrate that MURIM achieves up to 18% improvement in fairness metrics, reduces privacy attack success rates by 5-9%, and improves robustness against poisoning and noisy-gradient attacks by up to 85% compared to state-of-the-art baselines. Overall, MURIM effectively mitigates adversarial threats, promotes fair and truthful participation, and preserves stable model convergence across heterogeneous and dynamic federated settings.

Problem

• Addresses weak incentives and fairness in federated learning
• Mitigates privacy risks and adversarial attacks in collaborative training
• Enhances reliability and robustness against malicious or unreliable clients

Innovation

• Multidimensional reputation-based incentive mechanism for FL
• Jointly considers reliability, privacy, resource, and fairness
• Allocates incentives based on contribution, latency, and reputation