Summary of Flight: a Faas-based Framework For Complex and Hierarchical Federated Learning, by Nathaniel Hudson et al.
Flight: A FaaS-Based Framework for Complex and Hierarchical Federated Learning
by Nathaniel Hudson, Valerie Hayot-Sasson, Yadu Babuji, Matt Baughman, J. Gregory Pauloski, Ryan Chard, Ian Foster, Kyle Chard
First submitted to arxiv on: 24 Sep 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 The proposed Federated Learning (FL) framework, called Flight, is designed to support complex network topologies found in real-world distributed systems like the Internet-of-Things (IoT). Unlike existing frameworks that assume simple two-tier topologies, Flight can handle hierarchical multi-tier structures, asynchronous aggregation, and decouples control and data planes. The authors compare Flight’s performance with a state-of-the-art FL framework, Flower, showing improved scalability up to 2048 devices and reduced makespan across several models. Additionally, Flight’s hierarchical model reduces communication overheads by over 60%. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Flight is a new Federated Learning (FL) framework that works well in complex networks like the Internet-of-Things (IoT). Unlike other FL frameworks, which are simple and straightforward, Flight can handle many devices connected to each other in different ways. This makes it better than Flower, another popular FL framework, when there are many devices trying to learn at the same time. Flight is also faster and more efficient because it separates control and data planes. |
Keywords
» Artificial intelligence » Federated learning
