Summary of Flmarket: Enabling Privacy-preserved Pre-training Data Pricing For Federated Learning, by Zhenyu Wen et al.
FLMarket: Enabling Privacy-preserved Pre-training Data Pricing for Federated Learning
by Zhenyu Wen, Wanglei Feng, Di Wu, Haozhen Hu, Chang Xu, Bin Qian, Zhen Hong, Cong Wang, Shouling Ji
First submitted to arxiv on: 18 Nov 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Distributed, Parallel, and Cluster Computing (cs.DC)
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 Learning (FL) Market (FLMarket) tackles a crucial yet unexplored aspect of privacy-preserving machine learning: pre-training pricing. In contrast to existing approaches, FLMarket adopts an auction-based mechanism to reconcile the utility-privacy tradeoff. This innovative solution integrates a two-stage pricing framework with a security protocol, enabling accurate client selection for subsequent FL training. The experiments demonstrate that FLMarket achieves over 10% higher accuracy compared to state-of-the-art methods and outperforms in-training baselines by more than 2% in accuracy and 3x in runtime speed. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Federated Learning (FL) helps keep our data private, which is super important for things like healthcare and finance. Most research focuses on making FL better, but no one has looked at how to price the data before training starts. The new FLMarket solves this problem by using a special auction-based system that balances what’s useful with what’s private. This makes client selection more accurate, giving us higher accuracy gains of over 10% compared to other methods. Plus, it’s way faster! |
Keywords
* Artificial intelligence * Federated learning * Machine learning
