Summary of Locally Differentially Private Online Federated Learning with Correlated Noise, by Jiaojiao Zhang et al.
Locally Differentially Private Online Federated Learning With Correlated Noise
by Jiaojiao Zhang, Linglingzhi Zhu, Dominik Fay, Mikael Johansson
First submitted to arxiv on: 27 Nov 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (stat.ML)
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 proposes a novel locally differentially private (LDP) algorithm for online federated learning, which leverages temporally correlated noise to balance utility and privacy preservation. The authors develop a perturbed iterate analysis to mitigate the effects of correlated noise on local updates with non-IID streaming data. They also show how to manage drift errors from local updates for various nonconvex loss functions. The algorithm’s performance is bounded by a dynamic regret bound, which depends on key parameters and environmental changes. Experimental results validate the effectiveness of this approach. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper introduces a new way to learn together online while keeping individual data private. It works by adding random noise to updates from each device, making it harder for others to see what’s happening. The authors had to figure out how to make sure the noise doesn’t ruin the learning process and that local updates with different types of data don’t mess things up. They also found a way to deal with errors that happen when devices update their models differently. The algorithm works well even in situations where data changes frequently. |
Keywords
* Artificial intelligence * Federated learning
