Summary of Clients Collaborate: Flexible Differentially Private Federated Learning with Guaranteed Improvement Of Utility-privacy Trade-off, by Yuecheng Li et al.
Clients Collaborate: Flexible Differentially Private Federated Learning with Guaranteed Improvement of Utility-Privacy Trade-off
by Yuecheng Li, Tong Wang, Chuan Chen, Jian Lou, Bin Chen, Lei Yang, Zibin Zheng
First submitted to arxiv on: 10 Feb 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
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 introduces FedCEO, a novel federated learning framework that strikes a balance between model utility and user privacy in differential privacy. The framework uses efficient tensor low-rank proximal optimization to recover disrupted semantic information by smoothing the global semantic space for different privacy settings and continuous training processes. This approach improves the state-of-the-art (SOTA) utility-privacy trade-off bound by an order of √d, where d is the input dimension. The paper demonstrates its capabilities through experiments on representative image datasets, achieving significant performance improvements and strict privacy guarantees under different privacy settings. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps protect user data in federated learning by introducing a new way to balance model accuracy and privacy. It uses special math techniques to fix problems caused by adding noise to the model, which is needed to keep user data private. The new method, called FedCEO, can recover important information that was lost due to this noise. This makes it better than other methods at keeping both accurate models and private data. |
Keywords
* Artificial intelligence * Federated learning * Optimization
