Summary of Sharp Bounds For Sequential Federated Learning on Heterogeneous Data, by Yipeng Li and Xinchen Lyu
Sharp Bounds for Sequential Federated Learning on Heterogeneous Data
by Yipeng Li, Xinchen Lyu
First submitted to arxiv on: 2 May 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
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 paper explores Federated Learning (FL) paradigms, specifically Sequential FL (SFL), which trains models sequentially across clients. Unlike Parallel FL (PFL), SFL lacks convergence theory on heterogeneous data. To address this gap, the authors establish sharp convergence guarantees for SFL on heterogeneous data with both upper and lower bounds. The results show that SFL outperforms PFL in certain scenarios. Experimental validation is provided through quadratic functions and real datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Federated Learning is a way to train artificial intelligence models without collecting all the data in one place. There are two main ways to do this: parallel or sequential. In parallel, many computers work together at the same time, while in sequential, they work one after another. Researchers have been trying to figure out how well these methods work on different types of data, but there’s still a lot we don’t know about sequential learning on mixed data. To help answer this question, scientists developed new formulas that show when sequential learning works better than parallel and when it doesn’t. |
Keywords
» Artificial intelligence » Federated learning
