Summary of Federated Learning Architectures: a Performance Evaluation with Crop Yield Prediction Application, by Anwesha Mukherjee et al.
Federated Learning Architectures: A Performance Evaluation with Crop Yield Prediction Application
by Anwesha Mukherjee, Rajkumar Buyya
First submitted to arxiv on: 6 Aug 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 The proposed work presents a study on federated learning for crop yield prediction using Long Short-Term Memory Networks (LSTM). Two frameworks are implemented: centralized and decentralized. Centralized federated learning involves multiple clients sharing model updates with a server, which aggregates the information to create a global model. In contrast, the decentralized framework forms a collaborative network among devices using ring or mesh topologies, where each device receives updates from neighbors and performs aggregation. The performance of both frameworks is evaluated using metrics such as prediction accuracy, precision, recall, F1-score, and training time. The results show that centralized federated learning achieves ≥97% prediction accuracy, while decentralized federated learning achieves >97.5%. Additionally, the response time can be reduced by ~75% when using centralized federated learning compared to a cloud-only framework. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This study uses special technology called federated learning to predict how well crops will grow. It creates two ways for devices to work together: one where all the information is sent to a central server, and another where each device helps out its neighbors. The researchers tested these methods using complicated models like LSTMs. They found that both methods can be very accurate, with over 97% of predictions being correct. This technology could help farmers make better decisions about when to plant and how much water to use. |
Keywords
» Artificial intelligence » F1 score » Federated learning » Lstm » Precision » Recall
