A Blockchain-Based Dynamic Game Incentive Mechanism for Privacy-Preserving Computation

Privacy-preserving computation has become a critical component in data collaboration, yet existing systems often suffer from high cryptographic overhead and lack effective incentives for honest participation. To address these issues, we propose a blockchain-based dynamic game incentive mechanism for privacy-preserving computation, which uses a combination of rewards and regulatory penalties to incentivize rational participants, who aim to maximize their interests, to remain honest in secure multi-party computation (MPC). Building on this approach, we introduce a Shapley value-based reward distribution mechanism to ensure fair compensation according to each participant’s actual contribution. To maintain low on-chain cost while preventing manipulation, we introduce a seed-driven deterministic consensus procedure that enables off-chain Monte Carlo approximation of Shapley values while ensuring on-chain verifiability. The entire incentive and verification process is automated through smart contracts, ensuring transparency, verifiability, and trustless execution. Analysis shows that under rational assumptions, honest computation becomes the dominant strategy, prompting participants to proactively engage in honest computation to maximize their own gains. Our results provide a practical and incentive-compatible framework for fair, efficient, and trustworthy privacy-preserving computation in decentralized environments.