Summary of Fedmia: An Effective Membership Inference Attack Exploiting “all For One” Principle in Federated Learning, by Gongxi Zhu et al.
FedMIA: An Effective Membership Inference Attack Exploiting “All for One” Principle in Federated Learning
by Gongxi Zhu, Donghao Li, Hanlin Gu, Yuan Yao, Lixin Fan, Yuxing Han
First submitted to arxiv on: 9 Feb 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: 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 proposes a new approach to Membership Inference Attacks (MIAs) in Federated Learning (FL), addressing privacy concerns when training machine learning models on decentralized data. Existing methods focus on analyzing updates from the target client, but this approach underutilizes available information by also leveraging updates from non-target clients. The authors introduce FedMIA, a three-step MIA method that enhances effectiveness and outperforms existing MIAs in classification and generative tasks. FedMIA is robust against various defense strategies, Non-IID data, and different federated structures, making it a promising solution for FL practitioners. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps keep your personal information safe when you’re training artificial intelligence models on lots of different computers. There’s a problem called Membership Inference Attacks that tries to figure out where your information comes from. The authors came up with a new way to do this, using all the information from many computers instead of just one. It works better and is safer than other methods. This is important because it helps keep your data private when you’re working together with others. |
Keywords
* Artificial intelligence * Classification * Federated learning * Inference * Machine learning
