Summary of Federated Transformer: Multi-party Vertical Federated Learning on Practical Fuzzily Linked Data, by Zhaomin Wu et al.
Federated Transformer: Multi-Party Vertical Federated Learning on Practical Fuzzily Linked Data
by Zhaomin Wu, Junyi Hou, Yiqun Diao, Bingsheng He
First submitted to arxiv on: 23 Oct 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 the Federated Transformer (FeT), a novel framework for Vertical Federated Learning (VFL) that supports multi-party collaborations with fuzzy identifiers. The authors address the limitations of existing models by developing a new encoding method that incorporates transformer architecture distributed across parties, as well as a privacy framework integrating differential privacy and secure multi-party computation. Experimental results show FeT outperforms baseline models in terms of accuracy, achieving up to 46% improvement when scaled to 50 parties. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Imagine a world where different organizations can work together to train AI models without sharing their private data. This is the goal of Federated Learning (FL), and researchers have been working on making it more practical. One challenge is dealing with fuzzy identifiers, which are used to link different pieces of information from different parties. In this paper, the authors propose a new approach called the Federated Transformer (FeT) that can handle these fuzzy identifiers better than previous methods. They also developed a way to protect privacy while still sharing information. The results show that FeT is much more accurate and efficient than existing approaches. |
Keywords
» Artificial intelligence » Federated learning » Transformer
