分析联邦学习中玩家博弈与合作，提出监控和治理机制。
Gaming and Cooperation in Federated Learning: What Can Happen and How to Monitor It
- 构建分离良性合作与恶意操纵的分析框架
- 量化操纵性、博弈代价与合作代价，揭示规则影响
- 提供可审计的治理工具包，适合平台设计者使用
联邦学习（FL）的成功最终取决于参与者在部分可观测条件下的策略行为，但现有研究仍将其视为静态优化问题。本文将FL部署视为受策略驱动的系统，提出一个分析框架，区分福利提升行为与指标操纵行为。在此框架下，引入量化操纵性、博弈代价与合作代价的指标，研究规则、信息透明度、评估指标及聚合器切换政策如何重塑激励与合作模式。推导出遏制有害操纵同时保持良性合作的阈值条件，以及在预警信号达到临界时触发自动切换规则的机制。基于此，构建包含治理检查清单和具有可证明性能保证的审计预算分配算法的设计工具包。在多种模拟环境与真实联邦学习案例研究中，结果均与框架预测的定性和定量模式一致。整体成果为减少指标操纵、维持稳定高福利合作提供了设计原则与操作指南。
原文摘要 · Abstract (English)
The success of federated learning (FL) ultimately depends on how strategic participants behave under partial observability, yet most formulations still treat FL as a static optimization problem. We instead view FL deployments as governed strategic systems and develop an analytical framework that separates welfare-improving behavior from metric gaming. Within this framework, we introduce indices that quantify manipulability, the price of gaming, and the price of cooperation, and we use them to study how rules, information disclosure, evaluation metrics, and aggregator-switching policies reshape incentives and cooperation patterns. We derive threshold conditions for deterring harmful gaming while preserving benign cooperation, and for triggering auto-switch rules when early-warning indicators become critical. Building on these results, we construct a design toolkit including a governance checklist and a simple audit-budget allocation algorithm with a provable performance guarantee. Simulations across diverse stylized environments and a federated learning case study consistently match the qualitative and quantitative patterns predicted by our framework. Taken together, our results provide design principles and operational guidelines for reducing metric gaming while sustaining stable, high-welfare cooperation in FL platforms.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。