Summary of Federated Learning Client Pruning For Noisy Labels, by Mahdi Morafah et al.
Federated Learning Client Pruning for Noisy Labels
by Mahdi Morafah, Hojin Chang, Chen Chen, Bill Lin
First submitted to arxiv on: 11 Nov 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); 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 ClipFL framework addresses the challenge of noisy labels in Federated Learning (FL) by identifying and excluding noisy clients. This is achieved through a three-phase approach: pre-client pruning to calculate the Noise Candidacy Score (NCS), client pruning to exclude noisy clients, and post-client pruning for fine-tuning the global model. The framework outperforms state-of-the-art FL methods in terms of accuracy, convergence speed, and communication costs on diverse datasets with varying noise levels. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Federated Learning helps devices share knowledge without sharing data. But when devices have bad information, it can mess up the whole process. Scientists came up with a new way to deal with this problem called ClipFL. It looks at how well each device does and finds the ones that are wrong. Then, it gets rid of those bad devices and makes sure the good ones do their thing correctly. This helps make better decisions faster while also saving energy. |
Keywords
» Artificial intelligence » Federated learning » Fine tuning » Pruning
