Summary of Redefining Contributions: Shapley-driven Federated Learning, by Nurbek Tastan et al.
Redefining Contributions: Shapley-Driven Federated Learning
by Nurbek Tastan, Samar Fares, Toluwani Aremu, Samuel Horvath, Karthik Nandakumar
First submitted to arxiv on: 1 Jun 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 paper introduces ShapFed, a novel contribution assessment method for federated learning (FL) that uses Shapley values from cooperative game theory to provide a granular understanding of class-specific influences. The approach, called ShapFed-WA, outperforms conventional federated averaging in class-imbalanced scenarios and personalizes participant updates based on their contributions. Experiments on CIFAR-10, Chest X-Ray, and Fed-ISIC2019 datasets demonstrate the effectiveness of ShapFed in improving utility, efficiency, and fairness in FL systems. |
Low | GrooveSquid.com (original content) | Low Difficulty Summary Federated learning lets different people work together to train a big computer model without sharing all their data. But it’s hard to make sure everyone contributes equally and honestly. The proposed method, called ShapFed, helps by figuring out how much each person contributed to the final model. It works better than other methods in some situations and makes sure each person gets a different version of the model based on what they contributed. |
Keywords
» Artificial intelligence » Federated learning