Summary of Fedast: Federated Asynchronous Simultaneous Training, by Baris Askin et al.
FedAST: Federated Asynchronous Simultaneous Training
by Baris Askin, Pranay Sharma, Carlee Joe-Wong, Gauri Joshi
First submitted to arxiv on: 1 Jun 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Distributed, Parallel, and Cluster Computing (cs.DC)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
Summary difficulty | Written by | Summary |
---|---|---|
High | Paper authors | High Difficulty Summary Read the original abstract here |
Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper explores Federated Learning (FL), a technique enabling edge devices or clients to collaboratively train machine learning models without sharing private data. The authors focus on simultaneously training multiple FL models using a common set of clients, addressing the limitations of existing methods that employ synchronous aggregation, which can cause delays due to slow clients and large models. The proposed algorithm, FedAST, uses buffered asynchronous federated simultaneous training, adaptively allocating client resources across heterogeneous tasks. Theoretical convergence guarantees are provided for smooth non-convex objective functions. Experimental results demonstrate that FedAST outperforms existing methods, achieving a 46.0% reduction in time to train multiple tasks. |
Low | GrooveSquid.com (original content) | Low Difficulty Summary Federated Learning is a way for devices to work together on machine learning projects without sharing private information. Usually, this technique focuses on training one model at a time. This paper looks at how to train multiple models simultaneously using the same devices. They propose a new method called FedAST that helps speed up the process by avoiding slow devices and big models from holding everything back. The authors also prove that their method works well with certain types of problems. Their tests show that this new method is better than existing methods, taking 46% less time to finish multiple tasks. |
Keywords
» Artificial intelligence » Federated learning » Machine learning