发表机构
Curtin University; King Fahd University of Petroleum and Minerals; Umeå University（科廷大学; 法赫德国王石油矿产大学; 于默奥大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对FLaaS消费者可持续性约束导致的训练不稳定问题，提出SFLaaS框架，结合需求驱动搜索空间、碳可行性估计与调度策略及进化搜索，在实验中验证了方法的有效性。
AI 中文摘要
联邦学习即服务（FLaaS）消费者的可持续性约束，对维持FLaaS环境中碳可行的联邦训练构成重大挑战；这些约束常导致消费者参与不可行，且在严格碳约束下联邦训练不稳定。我们提出可持续联邦学习即服务（SFLaaS），这是一种针对异构可持续约束的碳约束神经架构搜索（NAS）框架。我们引入需求驱动的搜索空间，在联邦执行前将消费者可持续性概况转化为可行架构区域；开发消费者级碳可行性估计机制，在动态碳条件下评估候选架构；提出可持续消费者调度策略，自适应选择可行消费者并分配本地工作负载，以维持消费者参与和统计数据覆盖；采用进化搜索策略，在严格碳约束下联合优化预测性能、消费者可行性和参与覆盖。在真实世界数据集和模拟环境上的实验证明了所提方法的有效性。
英文摘要
The sustainability constraints of FLaaS consumers pose significant challenges to maintaining carbon-feasible federated training in FLaaS environments. These constraints often lead to infeasible consumer participation and unstable federated training under hard carbon constraints. We propose a Sustainable Federated Learning as a Service (SFLaaS), a carbon- constrained Neural Architecture Search (NAS) framework for heteroge- neous sustainable constraints. We introduce a requirement-driven search space that transforms consumer sustainability profiles into a feasible architecture region before federated execution. We develop a consumer-level carbon feasibility estimation mechanism to evaluate candidate architectures under dynamic carbon conditions. We propose a sustainable con- sumer scheduling strategy that adaptively selects feasible consumers and allocates local workloads to preserve consumer participation and statistical data coverage. An evolutionary search strategy jointly optimised for predictive performance, consumer feasibility, and participation coverage under hard carbon constraints. Experiments on real-world datasets and a simulated environment demonstrate the effectiveness of the proposed approach.
