Summary of Communication-efficient Federated Learning Over Wireless Channels Via Gradient Sketching, by Vineet Sunil Gattani et al.
Communication-Efficient Federated Learning over Wireless Channels via Gradient Sketching
by Vineet Sunil Gattani, Junshan Zhang, Gautam Dasarathy
First submitted to arxiv on: 30 Oct 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 proposed Federated Proximal Sketching (FPS) algorithm tackles major challenges in large-scale federated learning over wireless multiple access channels (MACs), including limited bandwidth, noisy communications, and heterogeneous datasets. FPS uses a count sketch data structure to efficiently compress data while maintaining accuracy, and adapts the loss function to handle varying degrees of data heterogeneity. The algorithm is theoretically proven to converge under mild conditions and outperforms state-of-the-art methods in synthetic and real-world experiments. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Federated learning over wireless multiple access channels (MACs) is important for many applications, but it’s hard because devices have limited bandwidth, send signals that can be noisy or wrong, and collect different types of data. To fix this, researchers created an algorithm called Federated Proximal Sketching (FPS). FPS uses a special way to store data to make it more efficient and handle the different kinds of data. They showed that FPS works well in practice by testing it on fake and real data. |
Keywords
» Artificial intelligence » Federated learning » Loss function
