| Title: |
FedASMU: Efficient Asynchronous Federated Learning with Dynamic Staleness-Aware Model Update |
| Authors: |
Liu, Ji; Jia, Juncheng; Che, Tianshi; Huo, Chao; Ren, Jiaxiang; Zhou, Yang; Dai, Huaiyu; Dou, Dejing |
| Source: |
Proceedings of the AAAI Conference on Artificial Intelligence; Vol. 38 No. 12: AAAI-24 Technical Tracks 12; 13900-13908 ; 2374-3468 ; 2159-5399 |
| Publisher Information: |
Association for the Advancement of Artificial Intelligence |
| Publication Year: |
2024 |
| Collection: |
Association for the Advancement of Artificial Intelligence: AAAI Publications |
| Subject Terms: |
ML: Distributed Machine Learning & Federated Learning; DMKM: Scalability; Parallel & Distributed Systems; ML: Scalability of ML Systems |
| Description: |
As a promising approach to deal with distributed data, Federated Learning (FL) achieves major advancements in recent years. FL enables collaborative model training by exploiting the raw data dispersed in multiple edge devices. However, the data is generally non-independent and identically distributed, i.e., statistical heterogeneity, and the edge devices significantly differ in terms of both computation and communication capacity, i.e., system heterogeneity. The statistical heterogeneity leads to severe accuracy degradation while the system heterogeneity significantly prolongs the training process. In order to address the heterogeneity issue, we propose an Asynchronous Staleness-aware Model Update FL framework, i.e., FedASMU, with two novel methods. First, we propose an asynchronous FL system model with a dynamical model aggregation method between updated local models and the global model on the server for superior accuracy and high efficiency. Then, we propose an adaptive local model adjustment method by aggregating the fresh global model with local models on devices to further improve the accuracy. Extensive experimentation with 6 models and 5 public datasets demonstrates that FedASMU significantly outperforms baseline approaches in terms of accuracy (0.60% to 23.90% higher) and efficiency (3.54% to 97.98% faster). |
| Document Type: |
article in journal/newspaper |
| File Description: |
application/pdf |
| Language: |
English |
| Relation: |
https://ojs.aaai.org/index.php/AAAI/article/view/29297/30446; https://ojs.aaai.org/index.php/AAAI/article/view/29297/30447; https://ojs.aaai.org/index.php/AAAI/article/view/29297 |
| DOI: |
10.1609/aaai.v38i12.29297 |
| Availability: |
https://ojs.aaai.org/index.php/AAAI/article/view/29297; https://doi.org/10.1609/aaai.v38i12.29297 |
| Rights: |
Copyright (c) 2024 Association for the Advancement of Artificial Intelligence |
| Accession Number: |
edsbas.7513B9AB |
| Database: |
BASE |