Summary of Fedtgp: Trainable Global Prototypes with Adaptive-margin-enhanced Contrastive Learning For Data and Model Heterogeneity in Federated Learning, by Jianqing Zhang et al.
FedTGP: Trainable Global Prototypes with Adaptive-Margin-Enhanced Contrastive Learning for Data and Model Heterogeneity in Federated Learning
by Jianqing Zhang, Yang Liu, Yang Hua, Jian Cao
First submitted to arxiv on: 6 Jan 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Cryptography and Security (cs.CR); 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 In a recent breakthrough in heterogeneous federated learning, researchers proposed prototype-based methods that reduce communication costs by sharing class representatives (prototypes) among clients. However, these prototypes are aggregated naively on the server using weighted averaging, leading to suboptimal global knowledge and poor performance. To overcome this challenge, the authors introduce FedTGP, an innovative approach that leverages Adaptive-margin-enhanced Contrastive Learning (ACL) to learn Trainable Global Prototypes (TGP). This novel method enhances prototype separability while preserving semantic meaning, outperforming state-of-the-art methods by up to 9.08% in accuracy while maintaining communication and privacy benefits. The code is available at this GitHub URL. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Recently, a new way of sharing information between different devices has been developed. This method is called Heterogeneous Federated Learning (HtFL). It allows for the sharing of information without having to share all the details. However, there was an issue with this method that needed to be solved. The issue was that the information being shared wasn’t very good because it was not being combined in the best way possible. To fix this problem, a new approach called FedTGP was developed. It uses something called Adaptive-margin-enhanced Contrastive Learning (ACL) to make sure the information is being combined correctly. This new method has been shown to work better than previous methods and can be used for things like improving AI models. |
Keywords
* Artificial intelligence * Federated learning
