Incentivizing Inclusive Contributions in Model Sharing Markets

2025-05-05

AI Summary

Decentralized private data is difficult to collaboratively utilize, and heterogeneous objectives among data owners lead to low participation incentives. Method: This paper proposes a graph-optimized, privacy-preserving model-sharing market framework for federated learning (FL). It is the first to integrate individual rationality and truthfulness incentives into FL, combining graph-structured model training, game-theoretic incentive design, personalized FL, and differential privacy–enhanced protocols. Contribution/Results: We establish the first personalized FL market that simultaneously ensures economic sustainability and privacy security. Our framework provides theoretical guarantees for voluntary and truthful participation. Extensive evaluation across 11 AI tasks—including LLM instruction fine-tuning—demonstrates optimal economic utility and model performance competitive with or superior to state-of-the-art baselines.

Technology Category

• Machine Learning: Privacy
• Game Theory and Economic Paradigms: Mechanism Design
• Search and Optimization: Learning to Search

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Economics, Online Markets and Human Computation: Fairness, privacy, and diversity in economic environments
• Security and Privacy: Privacy-enhancing technologies

Abstract

While data plays a crucial role in training contemporary AI models, it is acknowledged that valuable public data will be exhausted in a few years, directing the world's attention towards the massive decentralized private data. However, the privacy-sensitive nature of raw data and lack of incentive mechanism prevent these valuable data from being fully exploited. Addressing these challenges, this paper proposes inclusive and incentivized personalized federated learning (iPFL), which incentivizes data holders with diverse purposes to collaboratively train personalized models without revealing raw data. iPFL constructs a model-sharing market by solving a graph-based training optimization and incorporates an incentive mechanism based on game theory principles. Theoretical analysis shows that iPFL adheres to two key incentive properties: individual rationality and truthfulness. Empirical studies on eleven AI tasks (e.g., large language models' instruction-following tasks) demonstrate that iPFL consistently achieves the highest economic utility, and better or comparable model performance compared to baseline methods. We anticipate that our iPFL can serve as a valuable technique for boosting future AI models on decentralized private data while making everyone satisfied.

Problem

• Incentivizing diverse data holders to share private data
• Training personalized models without revealing raw data
• Ensuring individual rationality and truthfulness in incentives

Innovation

• Incentivized personalized federated learning (iPFL)
• Graph-based training optimization
• Game theory-based incentive mechanism