Summary of Fractional Order Distributed Optimization, by Andrei Lixandru et al.
Fractional Order Distributed Optimization
by Andrei Lixandru, Marcel van Gerven, Sergio Pequito
First submitted to arxiv on: 3 Dec 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 A novel distributed optimization framework, fractional order distributed optimization (FrODO), is proposed for modern machine learning applications like federated learning. FrODO incorporates fractional-order memory terms to improve convergence properties in challenging optimization landscapes, achieving provable linear convergence for any strongly connected network. Empirical results show that FrODO outperforms baselines by up to 4 times on ill-conditioned problems and 2-3 times in federated neural network training, while maintaining stability and theoretical guarantees. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary FrODO is a new way to optimize machine learning models when data is spread across different devices or networks. It helps make sure the optimization process works well even when the problem is tricky. FrODO was tested on some examples and showed it can be up to 4 times faster than other methods on hard problems, and 2-3 times faster on training neural networks in a distributed way. |
Keywords
» Artificial intelligence » Federated learning » Machine learning » Neural network » Optimization
