Summary of A Potential Game Perspective in Federated Learning, by Kang Liu et al.
A Potential Game Perspective in Federated Learning
by Kang Liu, Ziqi Wang, Enrique Zuazua
First submitted to arxiv on: 18 Nov 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
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 this study, researchers propose a new framework for federated learning (FL) where clients make their own decisions about how much effort to put into training machine learning models. This is different from traditional FL approaches where a central server tells clients what to do. The authors create a “game” model where each client’s payoff depends on the efforts they choose and the rewards they receive from the server. They show that there are certain situations where all clients will agree to put in the same amount of effort, and they identify an optimal reward factor for the server to use. The authors also test their approach using a real-world FL scenario and find that it works well. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Federated learning is a way for many devices or computers to work together to train machine learning models without sharing all their data. In this study, researchers want to know if each device can make its own decisions about how much effort to put into training the model. They create a game where each device gets a reward based on how well it does and how well the other devices do. The authors find that there is an optimal way for the devices to work together to get the best results, and they test this approach using real data. |
Keywords
* Artificial intelligence * Federated learning * Machine learning
