提出多模型联邦学习框架，动态调整批量大小并智能分配训练任务。
FLAMMABLE: A Multi-Model Federated Learning Framework with Multi-Model Engagement and Adaptive Batch Sizes
- 根据客户端能力动态调整批量大小，同时选择多个模型协同训练。
- 在多个数据集上提升训练速度1.1至10倍，最终准确率提高1.3%~5.4%。
- 适用于资源异构的分布式场景，适合研究联邦学习优化的开发者。
多模型联邦学习（MMFL）是联邦学习（FL）的新兴方向，允许多个模型在不同数据集上并行训练。在该设置下，优化模型准确率和训练时间需应对客户端间的数据与系统异构性，而模型间的额外异构性使挑战加剧。现有方案及单模型联邦学习的简单扩展均无法有效应对。为此，我们提出FLAMMABLE，一个完整的多模型联邦学习训练框架。其通过智能调节客户端批量大小，并在每轮训练中根据其系统能力动态选择多个特定模型进行训练，实现高效优化。为评估该框架，我们构建了首个针对MMFL的基准平台，以支持未来可复现的研究。在多个数据集与模型上的大量实验表明，相比若干已知基线，FLAMMABLE将MMFL的时-准性能提升1.1~10.0倍，最终模型准确率提高1.3%~5.4%。
原文摘要 · Abstract (English)
Multi-Model Federated Learning (MMFL) is an emerging direction in Federated Learning (FL) where multiple models are trained in parallel, generally on various datasets. Optimizing the models' accuracies and training times in the MMFL setting requires adapting to data and system heterogeneity across clients as in single-model FL; these challenges are amplified in the MMFL setting due to additional heterogeneity across models. Neither existing solutions nor naïve extensions of single-model FL frameworks efficiently address these challenges. To bridge this gap, we propose FLAMMABLE, a comprehensive MMFL training framework. FLAMMABLE optimizes model training by intelligently adapting client batch sizes while engaging them to train multiple carefully chosen models, depending on their system capabilities, in each training round. To evaluate FLAMMABLE, we develop the first benchmark platform for the MMFL setting, which may enable future reproducible MMFL research. Extensive evaluations on multiple datasets and models show that FLAMMABLE boosts the MMFL time-to-accuracy performance by 1.1$\sim$10.0$\times$ while improving the final model accuracy by 1.3$\sim$5.4\% compared to several known baselines.
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。