Summary of Privacy-preserving Federated Learning Via Dataset Distillation, by Shimao Xu et al.
Privacy-Preserving Federated Learning via Dataset Distillation
by ShiMao Xu, Xiaopeng Ke, Xing Su, Shucheng Li, Hao Wu, Sheng Zhong, Fengyuan Xu
First submitted to arxiv on: 25 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 This paper proposes Federated Learning for Information Protection (FLiP), which aims to bring the principle of least privilege (PoLP) to federated learning training. By applying elaborate information reduction on training data through local-global dataset distillation, FLiP minimizes the shared knowledge according to user intention while maintaining high model accuracy. The paper evaluates privacy performance using attribute inference and membership inference attacks, demonstrating that FLiP strikes a good balance between model accuracy and privacy protection. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Federated learning helps people share knowledge without sharing their personal data. But sometimes, people might not want to share all their information. This paper proposes a new way to train models with high accuracy while keeping sensitive info private. They call it FLiP, which brings the idea of “least privilege” (only sharing what’s necessary) to federated learning. By reducing the amount of data shared, FLiP keeps people’s privacy safe. |
Keywords
* Artificial intelligence * Distillation * Federated learning * Inference
