FLAMMABLE: A Multi-Model Federated Learning Framework with Multi-Model Engagement and Adaptive Batch Sizes
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866912644037869568 |
|---|---|
| author | Lin, Shouxu Pan, Zimeng Yao, Yuhang Noh, Haeyoung Zhang, Pei Joe-Wong, Carlee |
| author_facet | Lin, Shouxu Pan, Zimeng Yao, Yuhang Noh, Haeyoung Zhang, Pei Joe-Wong, Carlee |
| contents | 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$5.4\% compared to several known baselines. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_10380 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | FLAMMABLE: A Multi-Model Federated Learning Framework with Multi-Model Engagement and Adaptive Batch Sizes Lin, Shouxu Pan, Zimeng Yao, Yuhang Noh, Haeyoung Zhang, Pei Joe-Wong, Carlee Distributed, Parallel, and Cluster Computing Machine Learning Multi-Model Federated Learning (MMFL) is an emerging direction in Federated Learning (FL) where multiple models are trained in parallel, generally on various datasets. Optimizing the models' accuracies and training times in the MMFL setting requires adapting to data and system heterogeneity across clients as in single-model FL; these challenges are amplified in the MMFL setting due to additional heterogeneity across models. Neither existing solutions nor naïve extensions of single-model FL frameworks efficiently address these challenges. To bridge this gap, we propose FLAMMABLE, a comprehensive MMFL training framework. FLAMMABLE optimizes model training by intelligently adapting client batch sizes while engaging them to train multiple carefully chosen models, depending on their system capabilities, in each training round. To evaluate FLAMMABLE, we develop the first benchmark platform for the MMFL setting, which may enable future reproducible MMFL research. Extensive evaluations on multiple datasets and models show that FLAMMABLE boosts the MMFL time-to-accuracy performance by 1.1$\sim$10.0$\times$ while improving the final model accuracy by 1.3$\sim$5.4\% compared to several known baselines. |
| title | FLAMMABLE: A Multi-Model Federated Learning Framework with Multi-Model Engagement and Adaptive Batch Sizes |
| topic | Distributed, Parallel, and Cluster Computing Machine Learning |
| url | https://arxiv.org/abs/2510.10380 |