发表机构
Cognifinity（科格尼芬蒂）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究针对集群联邦学习的联盟形成问题，提出享乐势博弈方法，设计可转移剩余模型与分配规则，保证稳定划分与预算可行，在CIFAR-10实验中实现福利最优且性能可靠。
AI 中文摘要
集群联邦学习的优势在于将异构参与者组织成联盟，以训练联盟专属模型，但此类集群仅在参与者偏好其分配的联盟且所需转移可负担时才可持续。我们开发了一种可转移剩余模型，将学习收益、系统成本、参与者成本与货币转移相分离；一种分配规则将联盟剩余转化为享乐偏好，弱预算可行性保证协调者剩余非负。对于对称成对分配，诱导博弈为精确势博弈：存在纳什稳定划分，每一个严格最优反应过程均收敛，且在接受目的地同意的情况下，最优反应可达个体稳定划分。我们刻画了有界成对激励的可行性，并当保留松弛为次模时，在多项式预言时间内验证指数级多的预算约束。将福利分解为参与者势与保留松弛，可加性与乘性价格无政府状态保证成立，后者渐近紧；仅在成对可表示类上，精确平衡才实现福利最优稳定性，而仅预算可行可能导致无界福利损失。全局势最大化等价于加权最大一致相关聚类，近似后稳定化可满足由保留松弛与负边质量决定的端到端福利界，该界可通过显式构造达到。在预注册的五种子CIFAR-10研究中，该机制在每个主要实例上均达到经认证的估计表福利最优，而均等剩余共享在三个实例上无纳什稳定结果，成对验证增益比梯度对齐能更可靠地给出成对符号。
英文摘要
Clustered federated learning benefits from organizing heterogeneous participants into coalitions that train coalition-specific models, but such clustering is sustainable only if participants prefer their assigned coalition and the required transfers are affordable. We develop a transferable-surplus model separating learning benefit, system cost, participant cost, and monetary transfers; an allocation rule converts coalition surplus into hedonic preferences, and weak budget feasibility guarantees nonnegative retained coordinator surplus. For symmetric pairwise allocations the induced game is an exact potential game: a Nash-stable partition exists, every strict better-response process converges, and with destination consent accepted better responses reach an individually stable partition. We characterize feasibility of bounded pair incentives and verify the exponentially many budget constraints in polynomial oracle time when retained slack is submodular. Decomposing welfare into participant potential and retained slack yields additive and multiplicative price-of-stability guarantees, the latter asymptotically tight; exact balance gives welfare-optimal stability only on the pairwise-representable class, and budget feasibility alone permits unbounded welfare loss. Global potential maximization equals weighted maximum-agreement correlation clustering, and approximation followed by stabilization satisfies an end-to-end welfare bound governed by retained slack and negative-edge mass, attained by an explicit construction. In a preregistered five-seed CIFAR-10 study the mechanism reaches the certified estimated-table welfare optimum on every primary instance, equal-surplus sharing has no Nash-stable outcome on three, and pairwise validation gain gives far more reliable pair signs than gradient alignment.
Comments30 pages, 5 figures. Substantially revised and extended version of arXiv:2101.09673 (OptLearnMAS workshop at AAMAS 2021). Code and data: https://doi.org/10.5281/zenodo.21428635
