Summary of Dfml: Decentralized Federated Mutual Learning, by Yasser H. Khalil et al.
DFML: Decentralized Federated Mutual Learning
by Yasser H. Khalil, Amir H. Estiri, Mahdi Beitollahi, Nader Asadi, Sobhan Hemati, Xu Li, Guojun Zhang, Xi Chen
First submitted to arxiv on: 2 Feb 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); 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 This paper proposes a Decentralized Federated Mutual Learning (DFML) framework that addresses challenges in real-world devices, such as communication bottlenecks and single points of failure. The proposed framework is serverless, allowing for heterogeneous models and avoiding reliance on public data. DFML handles model and data heterogeneity through mutual learning, which distills knowledge between clients, and cyclically varying the amount of supervision and distillation signals. Experimental results show that DFML outperforms prevalent baselines in convergence speed and global accuracy under various conditions. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary In this paper, researchers create a new way to learn on many devices without relying on a central server. This helps with communication problems and makes it harder for hackers to target just one spot. The new method is called Decentralized Federated Mutual Learning (DFML). DFML lets different devices have their own models and data, but still shares knowledge to make them all better. Tests show that DFML works really well, especially when the data is mixed up or changing. |
Keywords
* Artificial intelligence * Distillation
