Summary of A New Perspective to Boost Performance Fairness For Medical Federated Learning, by Yunlu Yan et al.
A New Perspective to Boost Performance Fairness for Medical Federated Learning
by Yunlu Yan, Lei Zhu, Yuexiang Li, Xinxing Xu, Rick Siow Mong Goh, Yong Liu, Salman Khan, Chun-Mei Feng
First submitted to arxiv on: 12 Oct 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Cryptography and Security (cs.CR); Computers and Society (cs.CY); Image and Video Processing (eess.IV)
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 research proposes a novel approach called Fed-LWR, which aims to improve the fairness of federated learning (FL) in medical applications by addressing domain shift among datasets from different hospitals. The existing fair FL methods neglect this critical aspect, leading to biased performance. Fed-LWR dynamically perceives the bias of the global model across all hospitals by estimating layer-wise differences in feature representations between local and global models. It then assigns higher weights to hospitals with larger differences to minimize global divergence and obtain a fairer global model. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This study is about making sure that when doctors work together to analyze medical images, everyone gets the same good results. Right now, different hospitals have their own ways of doing things, which can cause problems. The researchers came up with a new way called Fed-LWR to fix this issue. It looks at how much each hospital’s data is different from the others and gives more importance to the ones that are most different. This helps make sure everyone gets good results. |
Keywords
» Artificial intelligence » Federated learning
