Summary of Befl: Balancing Energy Consumption in Federated Learning For Mobile Edge Iot, by Zehao Ju and Tongquan Wei and Fuke Shen
BEFL: Balancing Energy Consumption in Federated Learning for Mobile Edge IoT
by Zehao Ju, Tongquan Wei, Fuke Shen
First submitted to arxiv on: 5 Dec 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: 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 A novel federated learning framework called BEFL is proposed to enhance global model accuracy while minimizing energy consumption and reducing energy usage disparities among devices in Mobile Edge IoT. The approach jointly optimizes three objectives using the Sequential Least Squares Programming (SLSQP) algorithm for communication resource allocation, a heuristic client selection algorithm combining cluster partitioning with utility-driven approaches, and offline imitation learning during pre-training. Online, a ranking-based reinforcement learning approach is adopted to further boost training efficiency. Experimental results show that BEFL improves global model accuracy by 1.6%, reduces energy consumption variance by 72.7%, and lowers total energy consumption by 28.2% compared to existing methods. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary In this paper, researchers developed a new way to train AI models on devices while being mindful of the limited battery life and varying device capabilities. The goal is to create an accurate global model while also conserving energy and reducing inequalities in energy usage among devices. They proposed a framework called BEFL that balances three goals: improving model accuracy, minimizing overall energy consumption, and reducing energy disparities. The approach uses advanced algorithms like SLSQP and client selection methods to optimize the training process. The results show that BEFL outperforms existing methods by improving accuracy, reducing energy usage inequalities, and lowering overall energy consumption. |
Keywords
» Artificial intelligence » Federated learning » Reinforcement learning
