Summary of Pfedgame — Decentralized Federated Learning Using Game Theory in Dynamic Topology, by Monik Raj Behera et al.
pFedGame – Decentralized Federated Learning using Game Theory in Dynamic Topology
by Monik Raj Behera, Suchetana Chakraborty
First submitted to arxiv on: 5 Oct 2024
Categories
- Main: Machine Learning (stat.ML)
- Secondary: Cryptography and Security (cs.CR); Computer Science and Game Theory (cs.GT); Machine Learning (cs.LG)
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 pFedGame algorithm tackles the challenges faced by conventional federated learning frameworks, including performance bottlenecks, data bias, poor model convergence, and vulnerability to attacks. This novel game theory-based approach is designed for decentralized federated learning on temporally dynamic networks, where participants collaborate without a central aggregation server. The algorithm involves two sequential steps: peer selection and a two-player constant sum cooperative game for optimal federated learning aggregation. Experimental results demonstrate promising accuracy higher than 70% for heterogeneous data. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Federated learning helps many devices learn together without sharing their data. But current methods have problems like slow performance, biased results, and are vulnerable to attacks. The pFedGame algorithm is a new way to solve these issues by using game theory. It’s designed for networks that change over time, where devices work together without a central server. The algorithm has two parts: finding the right partners and playing a game to get the best result. Tests show this approach works well, with accuracy over 70% even when dealing with different types of data. |
Keywords
» Artificial intelligence » Federated learning
