Summary of Tackling Noisy Clients in Federated Learning with End-to-end Label Correction, by Xuefeng Jiang et al.
Tackling Noisy Clients in Federated Learning with End-to-end Label Correction
by Xuefeng Jiang, Sheng Sun, Jia Li, Jingjing Xue, Runhan Li, Zhiyuan Wu, Gang Xu, Yuwei Wang, Min Liu
First submitted to arxiv on: 8 Aug 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI)
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 FedELC framework tackles the issue of complicated label noise in federated learning by introducing a two-stage approach. The first stage detects noisy clients with higher label noise rates, while the second stage corrects the labels of these clients’ data through an end-to-end label correction framework. This is achieved by learning possible ground-truth labels via back propagation. The framework demonstrates superior performance compared to its counterparts across different scenarios and improves the data quality of detected noisy clients’ local datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Federated learning helps keep private information safe, but what happens when client data is noisy? Imagine trying to train a model with incorrect information! To solve this problem, researchers created FedELC, a two-part approach. First, it finds “noisy” clients with bad data. Then, it corrects those labels by guessing the right answers. This makes the whole process better and safer. |
Keywords
» Artificial intelligence » Federated learning
