Design and Implementation of Lightweight Federated Learning Platform LightFL
(School of Applied Science, Beijing Information Science and Technology University, Beijing 100192, China)
Abstract
The existing Federated Learning platforms are generally complex and require in-depth learning to master their functions for users. To address this issue, a lightweight Federated Learning platform, LightFL, is developed with the aim of lowering the entry barrier of usage for Federated Learning frameworks. The platform supports simulations of both Independent and Identically Distributed and Non-Independent and Identically Distributed data, and implements classic Federated Learning algorithms such as FedAvg, FedProx, and MOON. Through a modular architecture and lightweight code design, users can easily engage in Federated Learning and experiments, and the platform is easy to deploy and maintain. Compared to other platforms, the platform can configure the Federated Learning environment with only a small amount of code to write for users. It simplifies the workflow, reduces the entry barrier of usage, and facilitates the further development of new Federated Learning algorithms.
Keywords
lightweight platform; Federated Learning; Independent and Identically Distributed; Non-Independent and Identically Distributed
References
- [1] BANABILAH S,ALOQAILY M,ALSAYED E,et al. Federated Learning Review: Fundamentals, Enabling Technologies, and Future Applications [J/OL].Information Processing & Management,2022,59(6):103061[2024-09-15].https://doi.org/10.1016/j.ipm.2022.103061.
- [2] Liu Yang Research on Federated Learning Method Supporting Data Sharing in Industrial Internet of Things Environment [D]. Nanning: Guangxi University, 2024
- [3] Wang Wenshuo Research on Key Security Technologies of Federated Learning for Smart Healthcare [D]. Yantai: Yantai University, 2024
- [4] Chen Bing Research and Implementation of Privacy Protection Technology for Smart Home Based on Federated Learning [D]. Guangzhou: Guangzhou University, 2022
- [5] MCMAHAN B,MOORE E,RAMAGE D,et al. Communication-Efficient Learning of Deep Networks from Decentralized Data [C]//Proceedings of the 20th International Conference on Artificial Intelligence and Statistics.New York:PMLR,2017:1273-1282.
- [6] LI T,SAHU A K,ZAHEER M,et al. Federated Optimization in Heterogeneous Networks [J].Proceedings of Machine Learning and Systems,2020,2:429-450.
- [7] LI Q B,HE B S,SONG D. Model-Contrastive Federated Learning [C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.Piscataway:IEEE,2021:10713-10722.
- [8] LECUN Y,BOTTOU L,BENGIO Y,et al. Gradient-Based Learning Applied to Document Recognition [J].Proceedings of the IEEE,1998,86(11):2278-2324.
- [9] DENG L. The Mnist Database of Handwritten Digit Images for Machine Learning Research [J].IEEE Signal Processing Magazine,2012,29(6):141-142.
- [10] HULL J J. A Database for Handwritten Text Recognition Research [J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1994,16(5):550-554.
Funding
2024 Innovation and Entrepreneurship Training Program for College Students at Beijing University of Information Science and Technology (S202411232607); Support Program for Young Backbone Teachers at Beijing University of Information Science and Technology (YBT202450)
About the authors
Liu Ruoxuan (2003-), male, Manchu, from Haidian, Beijing, undergraduate studies, research direction: Federated Learning; Gao Linghang (2004-), male, Han ethnicity, from Fuzhou, Fujian Province, undergraduate studies, research direction: Federated Learning; Xie Guoyun (2004-), male, Han ethnicity, from Ganzhou, Jiangxi Province, undergraduate studies, research direction: Federated Learning; Zhang Yufan (2004-), male, Han ethnicity, from Jiaozuo, Henan Province, is currently pursuing a bachelor's degree with a research focus on federated learning; Li Hongen (2004-), male, Han ethnicity, from Lianyungang, Jiangsu Province, is currently pursuing a bachelor's degree with a research focus on federated learning; Wu Wenyun (2000-), female, Han ethnicity, from Handan, Hebei Province, is currently pursuing a master's degree in federated learning; Corresponding author: Wang Can (1990-), male, Han ethnicity, from Jining, Shandong, associate professor, doctoral student, research direction: Federated Learning.